RESUMO O conteúdo de água no solo é um fator-chave na avaliação nas áreas agrícola e ecológica. O objetivo deste trabalho foi explorar a combinação de informações multisatélites para abordar a questão da baixa precisão na recuperação do conteúdo de água do solo a partir de informações de banda única e fonte única, e estabelecer um modelo para estimativa do conteúdo de água do solo baseado no método multiespectral simulado. Foi realizado um experimento de determinação hiperespectral de amostras de solo (Latossolos) com valores variáveis de teor de água. A reflectância de oito sensores de satélite multiespectral foi remapeada com base nas funções de resposta espectral. Os índices de vegetação (VIs) foram estabelecidos combinando todas as bandas de sensores disponíveis. Os VIs significativos e a reflectância da banda foram extraídos pelos métodos de coeficiente de correlação e análise de importância de dados out-of-bag (OOB), e em seguida modelos lineares e não lineares foram estabelecidos para a estimativa do conteúdo de água no solo. Correlação significativa foi obtida entre refletância multiespectral simulada, VIs e teor de água no solo. Os modelos não lineares apresentaram melhor desempenho que o modelo linear. A combinação do modelo OOB e Random Forest (OOB-RF) obteve a maior acurácia de predição, com R2 calibração e R2 predição de 0,852 e 0,834, respectivamente. De forma geral, verificou-se que o método de modelagem OOB-RF baseado em sensoriamento remoto multiespectral foi viável para estimar o teor de água no solo.

ABSTRACT The soil water content is a key evaluation factor in the agricultural and ecological fields. The objective of this study was to explore the combination of multi-satellite information to address the issue of low accuracy in soil water content retrieval from single-band, single-source information and to establish a model for the estimation of the soil water content based on a simulated multi-spectral method. An experiment involving the hyperspectral determination of soil samples (Oxisols) with varying values of the water content was conducted. The reflectance of eight multi-spectral satellite sensors was resampled based on the spectral response functions. The vegetation indices (VIs) were established by pairing all available sensor bands. The significant VIs and band reflectance were extracted using the correlation coefficient and out-of-bag (OOB) data importance analysis methods, and then linear models and nonlinear models were established for soil water content estimation. A significant correlation was achieved between the simulated multi-spectral reflectance, VIs, and soil water content. The nonlinear models had a better performance than the linear model. The combined OOB and random forest (OOB-RF) model achieved the highest prediction accuracy, with R2 calibration and R2 prediction values of 0.852 and 0.834, respectively. Overall, it was verified that the OOB-RF modeling method based on multi-spectral remote sensing was feasible for estimating the soil water content.

Abstract This study evaluates the effectiveness of Recurrent Neural Networks (RNNs) and Transformer-based models in predicting the Air Quality Index (AQI). Accurate AQI prediction is critical for mitigating the significant health impacts of air pollution and plays a vital role in public health protection and environmental management. The research compares traditional RNN models, including Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks, with advanced Transformer architectures. Data were collected from a weather station in Cuenca, Ecuador, focusing on key pollutants such as CO,NO2,O3, PM2.5, and SO2. Model performance was assessed using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and the Coefficient of Determination (R2). The findings reveal that the LSTM model achieved superior performance, with an R2 of 0.701, an RMSE of 0.087, and an MAE of 0.056, demonstrating superior capability in capturing temporal dependencies within complex datasets. Conversely, while Transformerbased models exhibited potential, they were less effective in handling intricate time-series data, resulting in comparatively lower accuracy. These results position the LSTM model as the most reliable approach for AQI prediction, offering an optimal balance between predictive accuracy and computational efficiency. This research contributes to improving AQI forecasting and underscores the importance of timely interventions to mitigate the harmful effects of air pollution.

Resumen Este estudio evalúa la eficacia de las redes neuronales recurrentes (RNN) y los modelos basados en transformadores para predecir el índice de calidad del aire (ICA). La investigación compara los modelos RNN tradicionales, incluidos los de memoria a corto y largo plazo (LSTM) y la unidad recurrente controlada (GRU), con arquitecturas avanzadas de transformadores. El estudio utiliza datos de una estación meteorológica en Cuenca, Ecuador, centrándose en contaminantes como CO,NO2,O3, PM2.5 y SO2. Para evaluar el rendimiento de los modelos, se utilizaron métricas clave como el error cuadrático medio (RMSE), el error absoluto medio (MAE) y el coeficiente de determinación (R2). Los resultados del estudio muestran que el modelo LSTM fue el más preciso, alcanzando un R2 de 0,701, un RMSE de 0,087 y un MAE de 0,056. Esto lo convierte en la mejor opción para capturar dependencias temporales en los datos de series temporales complejas. En comparación, los modelos basados en transformadores demostraron tener potencial, pero no lograron la misma precisión que los modelos LSTM, especialmente en datos temporales más complicados. El estudio concluye que el LSTM es más eficaz en la predicción del ICA, equilibrando tanto la precisión como la eficiencia computacional, o que podría ayudar en intervenciones para mitigar la contaminación del aire.

Driven by the necessity to improve blast performance regarding fragment size distribution in limestone mines, this paper introduces the prediction and measurement of blast fragmentation distribution through the modified Kuz-Ram (MKR) and Kuznetsov-Cunningham-Ouchterlony (KCO) models. Precise prediction and measurement of rock fragmentation resulting from blasting play a critical role in hard rock ore extraction. This is due to the substantial influence of post-blast ore size distribution on the efficiency of all subsequent downstream rock handling and processing operations. Additionally, employing secondary breakage of boulders, when necessary, leads to an unjustified rise in production costs. In this paper, the MKR and KCO models were utilised to identify the most appropriate model for application in Bisil and Simba quarries situated in Kajiado County, Kenya. Split-Desktop was used for blast fragment measurement by processing digital images obtained after blasting and computing the fragment size distribution of the blasted rock. As per the investigation, the recorded percentages of boulders generated at the sites ranged from 7% to 12.3% for the Simba quarry and from 11.2% to 23.9% for the Bisil quarry, respectively. Regression, correlation, and root mean square error (RMSE) were employed as the primary performance indicators. The study concluded that KCO outperformed MKR in accurately predicting blast fragment sizes in the examined quarries.

Selecting ground motion prediction models (GMPM) is crucial in assessing seismic hazard levels in a specific region. This study investigates the influence of ground motion model (GMM) selection on seismic hazard assessment in Johannesburg, South Africa. Most seismic events affecting Johannesburg are related to mining activities and water ingress into the underground mine voids. The most seismically active regions in South Africa include areas around Klerksdorp, Welkom, and Carletonville. These regions experience a significant number of earthquakes, with mining activities being a major factor contributing to seismic events. Ten ground motion prediction equations were selected from regions of similar tectonics setting and were assessed to see if they correspond to the available local data. Six of these were also compared to the ground motions produced by the earthquake of 14 August 2014, which occurred in the Orkney area, one of the mining areas in South Africa. Sensitivity analyses were also carried out to assess the influence of ground motion prediction equations (GMPE) on the seismic hazard of Johannesburg. The results indicate that the choice of GMPM significantly affects the estimation of seismic hazard levels in Johannesburg. Different GMM can lead to varying ground shaking intensity predictions, impacting risk assessments and engineering design considerations. The findings highlight the importance of carefully selecting appropriate GMM tailored to Johannesburg's specific geological and tectonic characteristics. Understanding how different models perform in this region is essential for accurate seismic risk mitigation strategies.

ABSTRACT Predicting the adhesive force between steel reinforcement and concrete is crucial as it influences stress distribution and the overall mechanical behavior of reinforced concrete. This study proposes a novel approach to enhance bond strength prediction using machine learning (ML) models optimized through Bayesian optimization (BO). A dataset comprising 401 beam tests with six key factors was used to train three distinct ML algorithms—Support Vector Regression (SVR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). The prediction models were first trained on the full dataset, with BO applied to fine-tune hyperparameters and improve accuracy. Among these models, the BO-XGBoost achieved the best performance, with an R2 of 0.74, MAE of 1.412 MPa, and RMSE of 1.516 MPa on the test set, and R2 = 0.80, MAE = 0.950 MPa, RMSE = 1.200 MPa on the training set. In addition, a simplified model was developed, incorporating only three critical variables—rebar thickness, reinforcement tensile strength, and concrete compressive capacity—to make the model more applicable in real-world engineering scenarios. To further interpret the model’s predictions, Shapley additive explanations (SHAP) were employed, revealing the specific influence of each variable on bond strength. This study demonstrates that the integration of ML with Bayesian optimization can significantly improve the accuracy of bond strength predictions, offering valuable insights for structural design optimization.

ABSTRACT This experimental study investigates the axial compression effect of rectangular pultruded glass fiber-reinforced polymer (P-GFRP) tubular column sections, examining the impact of width-to-thickness ratio (B/t), aspect ratio (H/B), and column height on their structural performance. A total of 27 GFRP columns were subjected to axial compression tests to evaluate their ultimate load and initial stiffness. The columns exhibited a uniform failure pattern, characterized by crushing, mid-section fractures, and longitudinal splitting at the corners. The results revealed a negative correlation between the ultimate load and the aspect ratio, as well as the width-to-thickness ratio. This study utilized advanced machine learning algorithms, namely Response Surface Methodology (RSM) and Artificial Neural Network (ANN), to develop predictive models for the ultimate load of GFRP columns. The RSM model achieved an R2 value of 0.8347, demonstrating good accuracy in predicting ultimate load. The ANN model outperformed the RSM model, with an R criterion exceeding 0.68807 across training, testing, and validation phases, showing a stronger correlation between experimental and predicted outcomes. This research establishes a framework for forecasting the mechanical properties of column sections.

In general, the effects of contaminants from the oil industry, related to the hydraulic fracturing technique and its consequences on ecosystems, were evaluated by alternative methods, such as the use of the QSAR methodology. This method implements the quantitative structure-activity relationship to determine the toxicity of chemicals and predict their endpoints in biological systems, such as the mean lethal concentration (LC50). To determine the toxicity of different groups of chemical compounds used in hydraulic fracturing, tests were carried out using three aquatic species: Pimephales promelas, Oncorhynchus mykiss and Oreochromis niloticus. The results show that QSAR toolbox can predict aquatic toxicity, using the log Kow hydrophobicity descriptor and through this it’s observed that, in the three aquatic species, aldrin exhibits a high degree of toxicity compared to the other chemicals. In this sense, the QSAR models developed could be used to anticipate toxicity profiles and evaluate the chemical hazard of the studied substances.

Abstract In this paper, it is presented a general optimization methodology for improving empirical models predicting Okumura-Hata, Cost-231, ECC-33, and Egli using Genetic and Differential Evolution Algorithms. The methodology was tested using georeferenced signal power samples from Uberlândia, Brazil, for a television channel operating at 569.142857 MHz. Each parameter of the model formulas was adjusted with a variable optimized by the algorithms. A significant innovation was the inclusion of an altitude parameter weighted by an optimized coefficient, which notably enhanced the prediction accuracy. The primary contribution of this work is the development of a set of analytical equations derived from the proposed methodology, eliminating the need for computational power to estimate path loss for the evaluated models in the area in question. The performance of these equations was assessed using the Mean Squared Error (MSE) metric, demonstrating improvements of up to 92.03% over standard models, contingent on the empirical model and optimization algorithm applied.

Resumo Este trabalho foca-se na simulação do comportamento de pavimentos utilizando o Método de Elementos Finitos (MEF) com o objetivo de prever o aparecimento das fissuras durante sua vida útil. O método desenvolvido pode ajudar no planejamento da manutenção no sentido de proporcionar segurança e conforto aos usuários. São considerados os efeitos das cargas de tráfego e das variações de temperatura. Ferramentas de inteligência artificial (IA) são adotadas para reduzir o tempo necessário para a interpretação dos resultados das simulações para o desenvolvimento de um sistema eficiente de gestão de pavimentos. O sistema de gerenciamento de pavimentos rígidos desenvolvido utiliza a técnica de Redes Neurais Artificiais (RNA) para a previsão da resposta do pavimento ao acumulo de fadiga e do comportamento do pavimento modelado com alto grau de precisão. As redes apresentaram erros quadrados de 4.5% e 3.6% para gradientes de temperatura positivos e negativos e correlação de dados de 95% e 97%, respectivamente. O índice de tensão SR obtido pelo modelo desenvolvido em ABAQUS é 0.614 e o valor fornecido pelo sistema de gerenciamento de pavimentos é 0.648. A metodologia desenvolvida foi aplicada a um trecho da rodovia BR-101/NE, entre os estados da Paraíba PR e Pernambuco PE, no Brasil. Comparando os resultados com os limites previstos pelo Programa Cooperativo Nacional de Pesquisa Rodoviária (NCHRP) para o tráfego do trecho de estudo, este pavimento necessitaria de manutenção semestral. Os valores máximos de tensão de tração por flexão foram obtidos combinando o gradiente positivo de temperatura com a carga do eixo de tráfego na borda durante os meses de verão (dezembro a março) devido aos maiores valores de gradiente térmico. focase foca se MEF (MEF útil usuários IA (IA RNA (RNA precisão 45 4 5 4.5 36 3 6 3.6 95 97 97% respectivamente 0614 0 614 0.61 0648 648 0.648 BR101/NE, BR101NE BRNE BR 101/NE, 101 NE BR-101/NE PE Brasil NCHRP (NCHRP estudo semestral dezembro março térmico 4. 3. 9 061 61 0.6 064 64 0.64 BR101 BR101/NE 101NE 101/NE 10 06 0. BR10 1 BR1

Abstract This work is concerned with the simulation of pavement behavior using the Finite Element Method (FEM) with the goal of predicting the appearance of cracks during its useful life. The developed method can help maintenance planning towards providing users with safety and comfort. The effects of both traffic loads and temperature variations are considered. Artificial Intelligence (AI) tools are adopted to reduce the time necessary for the interpretation of simulation results for the development of an efficient pavement management system. The developed rigid pavement management system uses the Artificial Neural Networks (ANN) technique for the prediction of both pavement response to fatigue accumulation and the behavior of the modeled pavement with a high degree of precision. The networks showed 4.5% and 3.6% square errors for positive and negative temperature gradients and 95% and 97% data correlation, respectively. The SR stress ratio obtained by the developed ABAQUS model is 0.614 and the value provided by the pavement management system is 0.648. The developed methodology was applied to a section of highway BR-101/NE, between the states of Paraiba PR and Pernambuco PE, in Brazil. Comparing the results with the limits provided by National Cooperative Highway Research Program (NCHRP) for traffic in the study, this pavement would require maintenance every six months. The maximum values of tensile stress due to bending were obtained by combining the positive temperature gradient with the traffic axis load at the edge during the summer months (December to March) due to the higher thermal gradient values. FEM (FEM life comfort considered AI (AI ANN (ANN precision 45 4 5 4.5 36 3 6 3.6 95 97 correlation respectively 0614 0 614 0.61 0648 648 0.648 BR101/NE, BR101NE BRNE BR 101/NE, 101 NE BR-101/NE PE Brazil NCHRP (NCHRP study December March 4. 3. 9 061 61 0.6 064 64 0.64 BR101 BR101/NE 101NE 101/NE 10 06 0. BR10 1 BR1

The increase in the average temperature of the Earth due to excess CO2 and other greenhouse gases has caused glaciers to melt and sea levels to rise. The burning of fossil fuels is one of the main causes of this problem, so the use of biofuels is important for mitigating this environmental impact. In Brazil, biodiesel is mixed with mineral; however, irregularities have been detected, reducing the expected positive effects. Therefore, this study proposes the use of multivariate curve resolution with alternating least squares (MCR ALS) to quantify the content of methyl palm oil biodiesel in diesel blends and determine the concentration of the adulterant automotive lubricating oil in the blend. The model for quantifying biodiesel in diesel obtained R2 = 0.9997, root mean square error of calibration (RMSEC) = 0.56% (v v-1) and prediction (RMSEP) = 0.65% (v v-1), while the determination of adulteration in the biodiesel-diesel blend showed RMSEC = 0.67% (v v-1), RMSEP = 0.81% (v v-1) and R2 = 0.9992. These results demonstrate the ability of the models developed to predict both the content and quantity of possible adulterants in blends of palm oil biodiesel and mineral diesel, contributing to the quality control of this fuel. CO rise problem impact Brazil however detected effects Therefore MCR ALS R 09997 0 9997 0.9997 (RMSEC 056 56 0.56 v v1 1 v-1 (RMSEP 065 65 0.65 v1, , biodieseldiesel 067 67 0.67 081 81 0.81 09992 9992 0.9992 fuel 0999 999 0.999 05 5 0.5 v- 06 6 0.6 08 8 0.8 099 99 0.99 0. 09 9 0.9

A direct, high-throughput and accurate method, multi-molecular infrared spectroscopic (MMIR) data fusion, was established for the detection of trace of estrone, diethylstilbestrol, betamethasone and prednisone acetate in peptide powders based on near-infrared and mid-infrared fusion datasets. Tri step MMIR with progressively improved resolution captured key fingerprint features of hormones in the regions of 3430-3250, 2950-2850, 1750-1550 and 1335-1050 cm-1 which were exclusively selected as the datasets. The back propagation neural network (BPNN) model with the best performance used a mid-level fusion strategy and achieved excellent root mean square error of prediction (RMSEP, 1.101 mg kg-1) and residual prediction deviation (RPD, 9.890). Compared to those of single dataset models, the RMSEP was reduced by more than 82.03%. This method had remarkable practical significance for the direct and accurate quality evaluation of peptide powder products with favorable limit of detection (LOD) but without pretreatment and chemical agents, verifying the feasibility of multiple analytical instruments for rapid and comprehensive detection of powdered foods. highthroughput high throughput multimolecular multi molecular (MMIR estrone diethylstilbestrol nearinfrared near midinfrared mid datasets 34303250, 34303250 3430 3250, 3250 3430-3250 29502850, 29502850 2950 2850, 2850 2950-2850 17501550 1750 1550 1750-155 13351050 1335 1050 1335-105 cm1 cm 1 cm- BPNN (BPNN midlevel level RMSEP, (RMSEP 1101 101 1.10 kg1 kg kg-1 RPD, RPD (RPD 9.890. 9890 9.890 . 9 890 9.890) models 8203 82 03 82.03% LOD (LOD agents foods 3430325 343 325 3430-325 2950285 295 285 2950-285 1750155 175 155 1750-15 1335105 133 105 1335-10 110 10 1.1 kg- 989 9.89 89 820 8 0 82.03 343032 34 32 3430-32 295028 29 28 2950-28 175015 17 15 1750-1 133510 13 1335-1 11 1. 98 9.8 82.0 34303 3 3430-3 29502 2 2950-2 17501 1750- 13351 1335- 9. 82. 3430- 2950-

To rapidly detect different types of methamphetamine in the field, a portable near infrared spectroscopy (NIR) method was used to model and analyze the collected samples. Transmission spectra of 359 methamphetamine crystal samples (content range of 7.5-99.4%) and 187 methamphetamine tablet samples (content range of 2.43-23.91%) were collected, first-order derivatives and the Savitaki-Golay smoothing method were applied for spectral preprocessing. The drug concentrations in the calibration and validation sample sets were analyzed using high-performance liquid chromatography. The partial least squares quantitative analysis models were built, and the quantitative models for methamphetamine crystals were 2.10, 2.12 and 2.19% for root mean square error of cross-validation (RMSECV), root mean square error of validation (RMSEV) and root mean square error of prediction (RMSEP), respectively; in contrast, RMSECV, RMSEV, and RMSEP of methamphetamine tablets were 0.78, 0.80, and 0.82%, respectively. The results of the internal cross-validation and the methodology review show that the measurements of the modeled results are consistent with the true values, and that the model has good stability and can be used for accurate field measurements. Therefore, real time sample analysis can be performed using NIR technology combined with chemometric models, which helps the police to determine the quality of drugs in a timely manner for case analysis. (NIR 35 content 7.599.4% 75994 7.5 99.4% 7 5 99 4 7.5-99.4% 18 2.4323.91% 2432391 2.43 23.91% 2 43 23 91 2.43-23.91% firstorder first order SavitakiGolay Savitaki Golay preprocessing highperformance high performance chromatography built 210 10 2.10 212 12 2.1 219 19 2.19 crossvalidation cross RMSECV , (RMSECV) RMSEV (RMSEV RMSEP, (RMSEP) respectively contrast 078 0 78 0.78 080 80 0.80 082 82 0.82% values Therefore 3 599 7.599.4 7599 75 7. 994 99.4 9 7.5-99.4 1 4323 2.4323.91 243239 243 2.4 2391 23.91 2.43-23.91 21 2. (RMSECV (RMSEP 07 0.7 08 8 0.8 0.82 59 7.599. 759 99. 7.5-99. 432 2.4323.9 24323 24 239 23.9 2.43-23.9 0. 7.599 7.5-99 2.4323. 2432 23. 2.43-23. 7.59 7.5-9 2.4323 2.43-23 7.5- 2.432 2.43-2 2.43-

Livestock farms in Kenya face pressure to increase productivity amid rising global population. Cattle farming dominates, but small to medium-sized farms struggle with cattle insemination. Currently, visual observation is used for heat detection, with farmers maintaining farm journals. Modern methods utilizing sensors to improve estrus prediction are time-consuming, costly and need constant internet connection. This research proposes a novel approach-the use of an on-controller machine learning algorithm- for estrus prediction in cattle. Motion and temperature data was collected from two zero-grazed multiparous Holstein Friesian cows in Kiambu County, Kenya for 11 months. The data was cleaned and stored. Movement intensity profiles were derived by root-mean-squaring directional accelerometer values and averaging this over time. Validation was performed by observing cow behavior for indicators such as restlessness, mounting, and vulva swelling, with farmer predictions documented in their records. The collected data was then used to train a machine learning algorithm, with several models tested, and a neural network emerged as the best fit. The TensorFlow library facilitated the implementation of the algorithm on a microcontroller, allowing for the development of an animal tag featuring the ML algorithm. Results demonstrated 83.9% sensitivity, 89.0% specificity and 89.5% accuracy in detecting estrus, compared to farmer's visual observation, which had only 37% sensitivity. These findings underscore the potential to integrate machine learning into Precision Livestock Farming for estrus prediction, with prediction occurring directly on the animal tag offline. This integration holds promise for farmers, notably heightened insemination success rates, without necessitating significant financial investment.

ABSTRACT The maximum crown area is an important variable in the estimation of the level of competition that affects the growth potential and productivity of species. Therefore, the objective of the present study was to adjust regression models to predict the maximum crown diameter for Pinus cooperi Blanco in the forest region of El Salto, in the state of Durango. In particular, 95 trees of different diameter categories and growing free of competition were used. Independently, three regression models were analyzed, simple linear, second-degree polynomial and potential, to predict the maximum canopy diameter as a function of normal diameter, total height, height at the base of the crown, age and total length of the crown. The results showed that the second-degree polynomial model with the inclusion of normal diameter as an independent variable presented the best fit in the prediction of the maximum crown diameter. In general, normal diameter was the most significant variable for the prediction of maximum crown diameter in Pinus cooperi. Based on the results, two forest productivity scenarios related to crown competition factor (CCF) can be identified; when the CCF is 100 % or higher, timber productivity is optimal and when it is less than 100 %, productivity is not optimal, which can be attributed to little or no competition and deficiency in land use..

RESUMO A área máxima do dossel é uma variável importante na estimativa do nível de competição que afeta o potencial de crescimento e produtividade das espécies. Portanto, o objetivo deste estudo foi ajustar modelos de regressão para prever o diâmetro máximo da copa de Pinus cooperi Blanco na região florestal de El Salto, no estado de Durango. Em particular, serão utilizadas 95 árvores de diferentes categorias de diâmetro, criando limites à concorrência. De forma independente, foram analisados três modelos de regressão, linear simples, polinomial de segundo grau e potencial, para determinação do diâmetro máximo da copa em função do diâmetro normal, altura total, altura na base da copa, idade e comprimento total da copa. a xícara Os resultados mostraram que o modelo polinomial de segundo grau com a inclusão do diâmetro normal como variável independente apresentou o melhor ajuste na predição do diâmetro máximo da coroa. De modo geral, o diâmetro normal foi a variável mais significativa para predizer o diâmetro máximo da copa em Pinus cooperi. A partir dos resultados podem ser identificados cenários de produtividade florestal relacionados ao FCC; Quando o FCC é 100 % ou superior, a produtividade da produção é ótima e quando é inferior a 100 %, a produtividade não é ótima e pode ser atribuída a esse tempo ou à falta de competição e à deficiência no uso da terra.

RESUMEN El área máxima de copa es una variable importante en la estimación del nivel de competencia que afecta al potencial de crecimiento y productividad de las especies. Por lo tanto, el objetivo del presente estudio fue ajustar modelos de regresión para predecir el diámetro máximo de copa para Pinus cooperi Blanco en la región forestal de El Salto, en el estado de Durango. Particularmente se utilizaron 95 árboles de diferentes categorías diamétricas y creciendo libres de competencia. De manera independiente, se analizaron tres modelos de regresión, lineal simple, polinómico de segundo grado y potencial, para predecir el diámetro máximo de copa en función del diámetro normal, altura total, altura a la base de la copa, edad y longitud total de la copa. Los resultados evidenciaron que el modelo polinómico de segundo grado con la inclusión del diámetro normal como variable independiente presentó el mejor ajuste en la predicción del diámetro máximo de copa. En general el diámetro normal fue la variable más significativa para la predicción del diámetro máximo de copa en Pinus cooperi. Con base a los resultados se pueden identificar dos escenarios de productividad del bosque relacionados con factor de competencia de copa (FCC), cuando este es del 100 % o superior, la productividad maderable es óptima y cuando es menor al 100 % la productividad no es óptima pudiendo estar atribuido a la poca o nula competencia y deficiencia en el uso del suelo.

SUMMARY: Stature is an important component of biological profiles for forensic identification. In forensic situations, non- long bones might be needed for estimating stature if long bones are absent or fragmented. This study aimed to develop multiple regression model equations for stature estimation using scapula lengths in a Thai population. The sample included 200 dry scapula (100 male and 100 female) from the Osteology Research and Training Center, Faculty of Medicine, Chiang Mai University, Chiang Mai. Seven measurements were taken from the scapula for stature estimation. The results revealed that the maximum breadth of the scapula (MBS) provided the most accurate stature prediction model for the correlation coefficient and standard error of estimation (SEE) for males (r = 0.65, SEE = 5.96 cm), females (r = 0.60, SEE = 5.32 cm), and combined sexes (r = 0.79, SEE = 5.82 cm). The best multiple regression models were as follows: male stature (cm) = 75.57 + 6.42 MBS + 0.30 x the length of the glenoid fossa - superior angle (LGS), with a SEE of 5.67, female stature (cm) = 71.65 + 0.24 x the maximum height of the scapula (MHS) + 8.16 MBS - 1.02 x the maximum breadth of the glenoid fossa (MBG), with a SEE of 4.96 cm, combined sex stature (cm) = 52.61 + 0.25 MHS + 5.34 MBS + 0.25 LGS, with a SEE of 5.25 cm. This result indicates that the scapula is important in estimating the stature of skeletal remains in forensic cases, especially in a Thai population when long bones are unavailable for stature estimation.

RESUMEN: La estatura es un componente importante de los perfiles biológicos para la identificación forense. En situaciones forenses, es posible que se necesiten huesos no largos para estimar la estatura si los huesos largos están ausentes o fragmentados. Este estudio tuvo como objetivo desarrollar ecuaciones de modelos de regresión múltiple para la estimación de la estatura utilizando longitudes de escápula en una población tailandesa. La muestra incluyó 200 escápulas secas (100 de hombres y 100 de mujeres) del Centro de Investigación y Capacitación en Osteología, Facultad de Medicina, Universidad de Chiang Mai, Chiang Mai. Se tomaron siete medidas de la escápula para la estimación de la estatura. Los resultados revelaron que la anchura máxima de la escápula (MBS) proporcionó el modelo de predicción de estatura más preciso para el coeficiente de correlación y el error estándar de estimación (SEE) para hombres (r = 0,65, SEE = 5,96 cm), mujeres (r = 0,60, SEE = 5,32 cm) y sexos combinados (r = 0,79, SEE = 5,82 cm). Los mejores modelos de regresión múltiple fueron los siguientes: Estatura en hombres (cm) = 75,57 + 6,42 MBS + 0,30 x longitud de la fosa glenoidea - ángulo superior (LGS), con un SEE de 5,67, estatura en mujeres (cm) = 71,65 + 0,24 x altura máxima de la escápula (MHS) + 8,16 MBS - 1,02 x ancho máximo de la fosa glenoidea (MBG), con un SEE de 4,96 cm, estatura combinada de ambos sexos (cm) = 52,61 + 0,25 MHS + 5,34 MBS + 0,25 LGS, con un SEE de 5,25 cm. Este resultado indica que la escápula es importante para estimar la estatura de los restos óseos en casos forenses, especialmente en una población tailandesa en la que no se dispone de huesos largos para la estimación de la estatura.