Empirical Statistical Models Research Articles

A Multispectral Feature Selection Method Based on a Dual-Attention Network for the Accurate Estimation of Fractional Vegetation Cover in Winter Wheat

Spectral information plays a crucial role in fractional vegetation cover (FVC) estimation, and selecting the appropriate spectral information is essential for improving the accuracy of FVC estimation. Traditionally, spectral feature selection is primarily guided by physical mechanisms or empirical statistical models. This has led to the use of multispectral and hyperspectral images, which often result in missing or redundant information, thereby decreasing the efficiency and accuracy of FVC estimation. This study proposes a novel dual-attention network to select the feature bands of Sentinel-2 multispectral images for the accurate FVC estimation of winter wheat. In the first step, the importance of hyperspectral band reflectances was determined using simulated data from the PROSAIL model, by combining the dual-attention mechanism with the convolutional neural network (DAM-CNN). In the second step, the importance of Sentinel-2 multispectral bands was converted from the hyperspectral band importance identified in the previous stage, and subsequently ranked accordingly. Based on the feature ranking results, multispectral simulated data translated from hyperspectral simulated data were used for CNN training, and multispectral feature selection was conducted based on FVC accuracy. Finally, the selected features were assessed based on their performance in FVC estimation using a CNN model with real data. The experimental results indicate that during the key growth period of winter wheat, the combination of red, green, and red-edge bands significantly influences the FVC estimation accuracy. Band 3 (Green), band 4 (Red), band 5 (Red-edge 1), and band 6 (Red-edge 2) of Sentinel-2 satellite images contribute most significantly to winter wheat FVC estimation, achieving an accuracy comparable to that obtained using all bands, while reducing the training time by 19.1%, as confirmed by field survey data.

Precision Medicine With Data-driven Approaches: A Framework For Clinical Translation

Precision medicine-based approaches differentiate themselves by taking into consideration subpopulation variability (e.g. genetic variations, age, gender, race, addictions). To date, traditional models, such as Trial-and-Error Dosing, Empirical Treatment Guidelines, Statistical and Actuarial Models, Pathophysiological Models, and Clinical Judgment and Experience, have been generalized in healthcare fields. However, more comprehensive and innovative technologies are required besides these conventional modelings, which are subject to various limitations such as low efficiency and incapability of processing complex biological systems. Here we review diverse machine learning (ML) algorithms integrated with big data and omics and its applications in various aspects of precision medicine. ML is the branch of artificial intelligence (AI), which has been rapidly developed and highlighted as a promising method to decrease diagnostic errors and aid clinicians with decision-making in recent decades. We focused on applications of ML models such as support vector machine (SVM), K Nearest Neighbor (KNN) random forest (RF), convolutional neural networks (CNNs) and deep learning in drug toxicity prediction, cardiovascular diseases, neurodegenerative diseases, and cancer therapies within precision medicine and specific benefits and challenges of each. This review provides insights on the wider utilization in clinical environments by recognizing current advantages that are expected to expand the scope of AI-driven methods and issues that need to be addressed for further studies.

Artificial Intelligence Using FFNN Models for Computing Soil Complex Permittivity and Diesel Pollution Content

Soil pollution caused by hydrocarbons, such as diesel, poses significant risks to both human health and the ecosystem. The evaluation of soil pollution and various soil engineering applications often relies on the analysis of complex permittivity, encompassing parameters such as dielectric constant and dielectric loss. Various computational models, including theoretical physics-based models, mixture theory models, statistical empirical models, and artificial neural network (ANN) models, have been explored for computing soil complex permittivity and predicting water and pollutant content. Theoretical models require detailed data that is often unavailable, and thus have limited applicability. Mixture models tend to underestimate soil characteristics due to inaccuracies in permittivity estimation of soil phases. While empirical models are widely used, their applicability is restricted to specific soil types, datasets, and locations. ANN models offer promising predictions, accommodating nonlinear phenomena and allowing for missing information and variables. In this study, capacitive electromagnetic electrode sensors were utilized to determine the complex permittivity of soil contaminated with varying levels of diesel at different moisture levels. Theoretical mixture, empirical, and Feed Forward Neural Network (FFNN) models were employed to compute the permittivity of polluted soil based on its phases and to predict the level of diesel pollution. A comparison of these modeling approaches revealed that the FFNN model exhibited the best performance. The ANN model demonstrated superior performance metrics, including a high correlation coefficient and lower mean square error. Specifically, the correlation coefficients for the FFNN model were 0.9942 for training samples, 0.9967 for validation samples, and 0.9977 for test samples. Additionally, the ANN model yielded the lowest mean square error compared to the other three models. Doi: 10.28991/CEJ-2024-010-09-018 Full Text: PDF

Improved Winter Wheat Yield Estimation by Combining Remote Sensing Data, Machine Learning, and Phenological Metrics

Accurate yield prediction is essential for global food security and effective agricultural management. Traditional empirical statistical models and crop models face significant limitations, including high computational demands and dependency on high-resolution soil and daily weather data, that restrict their scalability across different temporal and spatial scales. Moreover, the lack of sufficient observational data further hinders the broad application of these methods. In this study, building on the SCYM method, we propose an integrated framework that combines crop models and machine learning techniques to optimize crop yield modeling methods and the selection of vegetation indices. We evaluated three commonly used vegetation indices and three widely applied ML techniques. Additionally, we assessed the impact of combining meteorological and phenological variables on yield estimation accuracy. The results indicated that the green chlorophyll vegetation index (GCVI) outperformed the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) in linear models, achieving an R2 of 0.31 and an RMSE of 396 kg/ha. Non-linear ML methods, particularly LightGBM, demonstrated superior performance, with an R2 of 0.42 and RMSE of 365 kg/ha for GCVI. The combination of GCVI with meteorological and phenological data provided the best results, with an R2 of 0.60 and an RMSE of 295 kg/ha. Our proposed framework significantly enhances the accuracy and efficiency of winter wheat yield estimation, supporting more effective agricultural management and policymaking.

STAR‐ESDM: A Generalizable Approach to Generating High‐Resolution Climate Projections Through Signal Decomposition

AbstractHigh‐resolution climate projections are critical to assessing climate risk and developing climate resilience strategies. However, they remain limited in quality, availability, and/or geographic coverage. The Seasonal Trends and Analysis of Residuals empirical statistical downscaling model (STAR‐ESDM) is a computationally‐efficient, flexible approach to generating such projections that can be applied globally using predictands and predictors sourced from weather stations, gridded data sets, satellites, reanalysis, and global or regional climate models. It uses signal processing combined with Fourier filtering and kernel density estimation techniques to decompose and smooth any quasi‐Gaussian time series, gridded or point‐based, into multi‐decadal long‐term means and/or trends; static and dynamic annual cycles; and probability distributions of daily variability. Long‐term predictor trends are bias‐corrected and predictor components used to map predictand components to future conditions. Components are then recombined for each station or grid cell to produce a continuous, high‐resolution bias‐corrected and downscaled time series at the spatial and temporal scale of the predictand time series. Comparing STAR‐ESDM output driven by coarse global climate model simulations with daily temperature and precipitation projections generated by a high‐resolution version of the same global model demonstrates it is capable of accurately reproducing projected changes for all but the most extreme temperature and precipitation values. For most continental areas, biases in 1‐in‐1000 hottest and coldest temperatures are <0.5°C and biases in the 1‐in‐1000 wet day precipitation amounts are <5 mm/day. As climate impacts intensify, STAR‐ESDM represents a significant advance in generating consistent high‐resolution projections to comprehensively assess climate risk and optimize resilience globally.

Predicting impacts of agricultural land use on stream and river biota: method review, evaluation, and guidance

Predicting land-use and land-management effects on stream and river biota is an important aspect of land-water management, yet there are no collations of what methods are available to carry out those assessments nor guidance on which methods to use. This paper summarises a range of methods with examples of their applications, comments on their strengths and weaknesses, evaluates them against a set of criteria, and provides guidance on method selection. Assessment methods include empirical statistical and mechanistic models, Bayesian networks, likelihood–consequence risk assessments, scoring methods, and hybrid methods, some of which can be informed by expert elicitation. An evaluation matrix for methods indicated that no single method is ideal, and selection of methods needs to carefully consider factors such as the physico-chemical stressor or biotic impact of interest, the intended stakeholders, and the scales of assessment. One emergent principle is the separation of relationships between land use and stressors from assessments of stressors and biota, for which alternative methods could be used. A tiered approach is recommended, whereby simple methods with low resource and time requirements are applied first, followed by more sophisticated methods for selected aspects if needed. There is a need for more ready-made methods at the screening level, as well as development of new methods to address remaining gaps such as multiple stressors.

A review of the satellite remote sensing techniques for assessment of runoff and sediment in soil erosion

Abstract Soil erosion monitoring is essential for the ecological evaluation and dynamic monitoring of land resources via remote sensing technology. In this paper, we provide new insights into the existing problems and development directions of traditional models, which are supported by new technologies. An important role is played by remote sensing information acquisition technology in the qualitative and quantitative evaluation of soil erosion, and the data and technical support provided are systematically reviewed. We provide a detailed overview of the research progress associated with empirical statistical models and physically driven process models of soil erosion, and the limitations of their application are also summarized. The preliminary integration of remote sensing data sources with high spatial and temporal resolution and new technologies for soil erosion monitoring enables the high-precision quantitative estimation of sediment transport trajectories, the watershed river network density, and the terrain slope, enhancing the accuracy of erosion factor identification, such as spectral feature recognition from erosion information, gully erosion feature extraction, and vegetation coverage estimation. However, the current erosion models, driven by algorithms and models, are not comprehensive enough, particularly in terms of the spatial feature extraction of erosion information, and there are limitations in the applicability and accurate estimation of such models.

Process Parameter Optimization of Directed Energy Deposited QT17‐4+ Steel

AbstractThe feasibility of using argon‐atomized QT 17‐4+ stainless steel powder for directed energy deposition (DED) additive manufacturing is studied. The QT 17‐4+ steel is a novel martensitic steel designed based on the compositional modification of the standard 17‐4 precipitation‐hardened (PH) stainless steel. This modification aims to achieve better mechanical properties of as‐deposited components compared to the heat‐treated wrought 17‐4PH steel. In this study, QT 17‐4+ steel powder is used for DED, for the first time. The influence of laser power, laser scan speed, powder feed rate, and hatch overlap on the density is studied. The central composite design is used to determine the experimental matrix of these factors. The response surface methodology is used to obtain the empirical statistical prediction model. Both columnar and equiaxed parent austenite grain structures are observed. X‐ray diffraction analyses reveal a decrease in the percentage of retained austenite from 19% in the powder to 5% after DED. The microhardness of the DED processed sample in the as‐deposited state is slightly higher than that of wrought 17‐4PH steel either solution‐annealed or H900‐aged. A higher 0.2% yield strength, a lower ultimate tensile strength, and lower elongation are observed for the vertically printed test sample, when compared to the horizontal one.

Modelling fuel oil transformation on geographically different seacoasts and assessing their self-cleansing capacity.

The present paper considers the results of long-term (up to 17years) in situ and laboratory research carried out on oiled French, Spanish, and Russian seacoasts. The objective of this research is to quantify the influence of geographical factors on the rates of natural transformation of the heavy fuel oil stranded ashore and to develop an empirical statistical model in order to evaluate the self-cleansing capacity of the coastal environment. In a number of field campaigns, 363 samples of weathered oil slicks and tar balls have been collected and analysed with the use of thin-layer chromatography combined with optical and gravimetric methods. The results obtained have been subjected to multiple nonlinear regression analyses. It has been shown that heavy fuel oil natural attenuation is more active in continental or estuarine environments influenced by nutrient-rich freshwater runoff and characterised by a higher number of sunny days, solar irradiation, and large temperature fluctuations. On the oceanic coasts, especially in sectors with low hydrodynamic energy, these processes take more time. The resulting model allows for the identification and mapping of the most vulnerable seacoasts, characterised by a low potential to degrade oil pollution. This information may be used in the contingency plans in order to optimise clean-up techniques and associated costs.

Service quality attributes affecting the satisfaction of KTM Komuter services

ABSTRACT This research aims to evaluate the service quality within the commuter system of Klang Valley, Malaysia, specifically the KTM Komuter, by constructing empirical statistical models known as KOMIQUAL models. The primary objective is to identify and quantify the significance or impact of various variables on commuter quality and service, focusing on the Port Klang-Sentul route in the year 2010. The empirical investigation conducted at fourteen commuter stations in Klang Valley was prompted by concerns about severe overcrowding of passengers on rail stations, platforms, and within rail coaches. In addition, some major stations were selected because they faced high demand for passengers. Valid questionnaire responses were statistically analyzed using factor analysis and the Structural Equation Modeling-Analysis of Moment Structures (SEM-AMOS) Graphics and software. Findings showed that the major service characteristics like parking facilities, train efficiency, people services, and space comfort had the highest positive effect on service quality. This research would help the operator to monitor and improve the service.

Modeling Shallow Landslide Runout Distance in Eocene Flysch Facies Using Empirical–Statistical Models (Western Black Sea Region of Türkiye)

Uncertainties related to runout distances in shallow landslide analyses may not only affect lives but may also result in economic losses. Owing to the increase in shallow landslides, which are especially triggered by heavy rainfall, runout distances have been investigated to decipher whether applications of a functional runout distance are feasible. This paper aims to give insights into the modeling of the shallow landslide runout probability in Eocene flysch facies in the Western Black Sea region of Türkiye. There are two main stages in this study—which are dominated by empirical models, the detection of initiation points, and propagation—which help us to understand and visualize the possible runout distances in the study area. Shallow landslide initiation point determination using machine learning has a critical role in the ordered tasks in this study. Modified Holmgren and simplified friction-limited model (SFLM) parameters were applied to provide a good approximation of runout distances during the propagation stage using Flow-R software. The empirical model parameters suggested for debris flows and shallow landslides were investigated comparatively. The runout distance models had approximately the same performance depending on the debris flow and shallow landslide parameters. While the impacted total runout areas for the debris flow parameters were predicted to amount to approximately 146 km2, the impacted total runout areas for the shallow landslide parameters were estimated to be about 101 km2. Considering the inclusion of the RCP 4.5 and RCP 8.5 precipitation scenarios in the analyses, this also shows that the shallow landslide and debris flow runout distance impact areas will decrease. The investigation of runout distance analyses and the inclusion of the RCP scenarios in the runout analyses are highly intriguing for landslide researchers.

Statistical characterization of specimen size and temperature dependence of cleavage fracture toughness of ferritic steels

AbstractBased on the numerical results on the non‐linear correlation between volume of plastic zone and global loading level represented by stress intensity factor, the dependence of cleavage fracture toughness of ferritic steels on specimen size and temperature is evaluated based on an empirical statistical model. The statistical correlation is supported by two sets of toughness data measured from different sized compact tension specimens at different temperatures. To further improve the accuracy of the statistical model, future work is identified to quantify the variation of volume of plastic deformation zone with loading level and measure the average volume for each microcrack at different temperature.

Synchronous Retrieval of Wheat Cab and LAI from UAV Remote Sensing: Application of the Optimized Estimation Inversion Framework

Chlorophyll a and b content (Cab) and leaf area index (LAI) are two key parameters of crops, and their quantitative inversions are important for growth monitoring and the field management of wheat. However, due to the close correlation between the spectral signals of these two parameters and the effects of soil and atmospheric conditions, as well as modeling errors, synchronous retrieval of LAI and Cab from remote sensing data is still a challenging task. In a previous study, we introduced the optimal estimation theory and established the inversion framework by coupling the PROSAIL (PROSPECT + SAIL) model with the unified linearized vector radiative transfer model (UNL-VRTM). The framework fully utilizes the simulated radiance spectra for synchronous retrieval of Cab and LAI at the UAV observation scale and has good convergence and self-consistency. In this study, based on this inversion framework, synchronized retrieval of Cab and LAI was carried out by real wheat UAV observation data and validated with the ground-measured data. By comparing with the empirical statistical model constructed by the PROSAIL model and coupled model, least squares support vector machine (LSSVM), and random forest (RF), the proposed method has the highest accuracy of Cab and LAI estimated from UAV multispectral data (for Cab, R2 = 0.835, RMSE = 14.357; for LAI, R2 = 0.892, RMSE = 0.564). Our proposed method enables the fast and efficient estimation of Cab and LAI in multispectral data without prior measurements and training.

A Statistical Empirical Model and RSM‐Guided Isolation and Characterization of Cellulose from Invasive Weed Senna didymobotrya (Fresen.) Irwin & Barneby, through Chemical and Spectroscopic Techniques

A Statistical Empirical Model and RSM‐Guided Isolation and Characterization of Cellulose from Invasive Weed <i>Senna didymobotrya</i> (Fresen.) Irwin &amp; Barneby, through Chemical and Spectroscopic Techniques

Predicting country-specific financing capacity for renewable energy project

This study aims to scrutinize the various determinants that influence a nation’s ability to fund and support renewable energy ventures, encompassing factors such as economic stability, regulatory environment, energy demand, and access to capital markets. By drawing on a range of empirical data, financial indicators, and statistical models, this study seeks to determine which factor most potent when predicting financing capacity of a specific country towards renewable. A secondary research using published data by government publications and non-governmental databases is the research method for the present study. The data derived from these databases organized into tables to allow for regression analysis to be conducted to achieve the research objectives. The results from the regression analysis indicate that stock market and inflation rate are significant variables should be included in the predictive model of financing capacity for renewable energy.

Improved methods for determining the leaf surface index using unmanned aerial vehicles to determine the productivity of phytocenoses

In landscape and biogeographic studies, the assessment of the photosynthesis process, which affects the possibility of determining the productivity of phytocenoses, calculating the growth of phytomass, is of particular relevance. The use of unmanned aerial vehicles (UAVs) to evaluate the leaf surface index (LAI) is gaining more and more scope over time due to their low price, high operational efficiency and accuracy. The basis for determining LAI is a model for the slit fraction, theoretically estimated using the Bera-Booger-Lambert law. The possibility of determining the LAI of vegetation using a UAV, on board of which either a lidar or a multispectrometer can be installed, is analyzed. In the first problem, LAI is determined by calculating the logarithm of the slit function (fraction) multiplied by the cosine of the scanning angle and divided by the attenuation coefficient. The second problem uses the existing correlations between known vegetation indices and LAI. Empirical statistical regression models may be suitable for determining LAI after determining various vegetation indices. Based on the results of multispectral measurements, it was found that the LAI determination technique based on measuring the intensity of rays that passed through the crown of plants leads to highly noisy estimates. For this reason, it was decided to use the slit fraction (GF), while using the experimentally established fact that when multiplying the logarithm of the vegetation index by the height of the crown, this correlation is significantly enhanced. To increase the reliability of the obtained LAI values, it is proposed to use the average integral value of this indicator, calculated by composing and calculating an optimization variational problem containing an additional restrictive condition. At the same time, it is possible to solve both problems to the maximum, i.e. it becomes possible to increase the signal-to-noise ratio of the calculated value of the LAI index. In both optimization calculation procedures, there are generalized indicators that have different physical meanings.

Digital measurement of SARS-CoV-2 transmission risk from 7 million contacts.

How likely is it to become infected by SARS-CoV-2 after being exposed? Almost everyone wondered about this question during the COVID-19 pandemic. Contact-tracing apps1,2 recorded measurements of proximity3 and duration between nearby smartphones. Contacts-individuals exposed to confirmed cases-were notified according to public health policies such as the 2 m, 15 min guideline4,5, despite limited evidence supporting this threshold. Here we analysed 7 million contacts notified by the National Health Service COVID-19 app6,7 in England and Wales to infer how app measurements translated to actual transmissions. Empirical metrics and statistical modelling showed a strong relation between app-computed risk scores and actual transmission probability. Longer exposures at greater distances had risk similar to that of shorter exposures at closer distances. The probability of transmission confirmed by a reported positive test increased initially linearly with duration of exposure (1.1% per hour) and continued increasing over several days. Whereas most exposures were short (median 0.7 h, interquartile range 0.4-1.6), transmissions typically resulted from exposures lasting between 1 h and several days (median 6 h, interquartile range 1.4-28). Households accounted for about 6% of contacts but 40% of transmissions. With sufficient preparation, privacy-preserving yet precise analyses of risk that would inform public health measures, based on digital contact tracing, could be performed within weeks of the emergence of a new pathogen.

Exploring the Nexus of Technology Availability, Child-Friendly Interface Design, Early Childhood Digital Literacy, Cognitive Skills, and Creativity in Language Learning in the Context of Banten Javanese Language

This research delves into intricate relationships involving technology availability, child-friendly interface design, early childhood digital literacy, cognitive skills, and creativity within the Banten Javanese language in Indonesia. Employing empirical analysis and statistical modelling, it examines the impact of technology and child-friendly interfaces on digital literacy, cognitive skills, and creativity, providing insights for theory and practice. The findings show significant positive correlations between technology availability, digital literacy, and cognitive skills among young learners. Child-friendly interfaces enhance digital literacy, cognitive abilities, and creativity in language learning. Theoretical contributions underscore integrating technology into language preservation, emphasizing its influence on digital literacy, cognitive development, and creativity in language learning. The study highlights digital literacy’s mediating role in the relationship between technology and language development. Educators and advocates can use these insights to design language programs prioritizing digital literacy and creative language acquisition. However, the study’s limitations, such as contextual specificity and a limited sample size, suggest the need for future research, including longitudinal and cross-cultural studies. The study’s novelty lies in its interdisciplinary approach, offering a nuanced understanding of how technology can enrich language acquisition while preserving cultural heritage.

Grain yield estimation of wheat-maize rotation cultivated land based on Sentinel-2 multi-spectral image: A case study in Caoxian County, Shandong, China

Establishing the remote sensing yield estimation model of wheat-maize rotation cultivated land can timely and accurately estimate the comprehensive grain yield. Taking the winter wheat-summer maize rotation cultivated land in Caoxian County, Shandong Province, as test object, using the Sentinel-2 images from 2018 to 2019, we compared the time-series feature classification based on QGIS platform and support vector machine algorithm to select the best method and extract sowing area of wheat-maize rotation cultivated land. Based on the correlation between wheat and maize vegetation index and the statistical yield, we screened the sensitive vegetation indices and their growth period, and obtained the vegetation index integral value of the sensitive spectral period by using the Newton-trapezoid integration method. We constructed the multiple linear regression and three machine learning (random forest, RF; neural network model, BP; support vector machine model, SVM) models based on the integral value combination to get the best and and optimized yield estimation model. The results showed that the accuracy rate of extracting wheat and maize sowing area based on time-series features using QGIS platform reached 94.6%, with the overall accuracy and Kappa coefficient were 5.9% and 0.12 higher than those of the support vector machine algorithm, respectively. The remote sensing yield estimation in sensitive spectral period was better than that in single growth period. The normalized differential vegetation index and ratio vegetation index integral group of wheat and enhanced vegetation index and structure intensify pigment vegetable index integral group of maize could more effectively aggregate spectral information. The optimal combination of vegetation index integral was difference, and the fitting accuracy of machine learning algorithm was higher than that of empirical statistical model. The optimal yield estimation model was the difference value group-random forest (DVG-RF) model of machine learning algorithm (R2=0.843, root mean square error=2.822 kg·hm-2), with a yield estimation accuracy of 93.4%. We explored the use of QGIS platform to extract the sowing area, and carried out a systematical case study on grain yield estimation method of wheat-maize rotation cultivated land. The established multi-vegetation index integral combination model was effective and feasible, which could improve accuracy and efficiency of yield estimation.

A PROSAIL model with a vegetation index lookup table optimized with in-situ statistics for rapeseed leaf area index estimation using diverse unmanned aerial vehicle sensors in the Yangtze River Basin

For winter rapeseed, timely monitoring of the leaf area index (LAI) near the winter solstice is critical for estimating its overwintering survival rate and yield. While unmanned aerial vehicle (UAV) remote sensing techniques have emerged as a promising means for crop LAI monitoring, many existing crop LAI models have only been evaluated using a single sensor system, a single cultivar, a single development stage, and/or a single site. As a result, the generalization capability of these LAI models is limited. In this study, four empirical statistical models (ESMs) and a new PROSAIL-based lookup table (LUT) method were developed to estimate the LAI of 24 winter rapeseed cultivars mainly grown in the Yangtze River Basin. The raw LUT was optimized using a vegetation index (VI) strategy after screening out inefficient parameter combinations using in-situ UAV reflectance and measured LAIs. The model transferability was evaluated across three years, six regional sites around Yangtze River Basin, different development stages, and two data sources from different sensors. An artificial neural network model in the ESMs performed best for data acquired in 2019 and 2020, but could not be used to estimate the LAI in 2018 due to the use of different sensors for data acquisition. In contrast, the optimized VI-LUT method was robust and outperformed the raw LUT for LAI retrieval with a root mean squared error of <0.7 for all data sets. These findings suggest that the optimized VI-LUT method has potential for estimating LAI at different spatial and temporal scales, especially for the situation where sensor types are difficult to be unified in multiple locations. Moreover, the genotype in these cultivars did not appear to dominate LAI generation before overwintering, but the climatic zone did. These findings have implications for the study of genotype-environment interactions in rapeseed cultivation.