Search

Considering the demand for the optimization of the technical factors of spraying for a greater area coverage and minimal drift, field tests were carried out to determine the interaction between the area coverage, number of droplets per cm2, droplet diameter, and drift. The studies were conducted with two different types of sprayers (axial and radial fan) in an apple orchard and a vineyard. The technical factors of the spraying interactions were nozzle type (ISO code 015, code 02, and code 03), working speed (6 and 8 km h−1), and spraying norm (250–400 L h−1). The airflow of both sprayers was adjusted to the plantation leaf mass and the working pressure was set for each repetition separately. A method using water-sensitive paper and a digital image analysis was used to collect data on coverage factors. The data from the field research were processed using four machine learning models: quantile random forest (QRF), support vector regression with radial basis function kernel (SVR), Bayesian Regularization for Feed-Forward Neural Networks (BRNN), and Ensemble Machine Learning (ENS). Nozzle type had the highest predictive value for the properties of number of droplets per cm2 (axial = 69.1%; radial = 66.0%), droplet diameter (axial = 30.6%; radial = 38.2%), and area coverage (axial = 24.6%; radial = 34.8%). Spraying norm had the greatest predictive value for area coverage (axial = 43.3%; radial = 26.9%) and drift (axial = 72.4%; radial = 62.3%). Greater coverage of the treated area and a greater number of droplets were achieved with the radial sprayer, as well as less drift. The accuracy of the machine learning model for the prediction of the treated surface showed a satisfactory accuracy for most properties (R2 = 0.694–0.984), except for the estimation of the droplet diameter for an axial sprayer (R2 = 0.437–0.503). [ABSTRACT FROM AUTHOR]

The research on lignocellulose pretreatments is generally performed through experiments that require substantial resources, are often time-consuming and are not always environmentally friendly. Therefore, researchers are developing computational methods which can minimize experimental procedures and save money. In this research, three machine learning methods, including Random Forest (RF), Extreme Gradient Boosting (XGB) and Support Vector Machine (SVM), as well as their ensembles were evaluated to predict acid-insoluble detergent lignin (AIDL) content in lignocellulose biomass. Three different types of harvest residue (maize stover, soybean straw and sunflower stalk) were first pretreated in a laboratory oven with hot air under two different temperatures (121 and 175 °C) at different duration (30 and 90 min) with the aim of disintegration of the lignocellulosic structure, i.e., delignification. Based on the leave-one-out cross-validation, the XGB resulted in the highest accuracy for all individual harvest residues, achieving the coefficient of determination (R2) in the range of 0.756–0.980. The relative variable importances for all individual harvest residues strongly suggested the dominant impact of pretreatment temperature in comparison to its duration. These findings proved the effectiveness of machine learning prediction in the optimization of lignocellulose pretreatment, leading to a more efficient lignin destabilization approach. [ABSTRACT FROM AUTHOR]

This study employed an ensemble machine learning approach to evaluate the effect of bioclimatic covariates on the prediction accuracy of soil total carbon (TC) in the Pannonian biogeoregion. The analysis involved two main segments: (1) evaluation of base environmental covariates, including surface reflectance, phenology, and derived covariates, compared to the addition of bioclimatic covariates; and (2) assessment of three individual machine learning methods, including random forest (RF), extreme gradient boosting (XGB), and support vector machine (SVM), as well as their ensemble for soil TC prediction. Among the evaluated machine learning methods, the ensemble approach resulted in the highest prediction accuracy overall, outperforming the individual models. The ensemble method with bioclimatic covariates achieved an R2 of 0.580 and an RMSE of 10.392, demonstrating its effectiveness in capturing complex relationships among environmental covariates. The results of this study suggest that the ensemble model consistently outperforms individual machine learning methods (RF, XGB, and SVM), and adding bioclimatic covariates improves the predictive performance of all methods. The study highlights the importance of integrating bioclimatic covariates when modeling environmental covariates and demonstrates the benefits of ensemble machine learning for the geospatial prediction of soil TC. [ABSTRACT FROM AUTHOR]

The increasing wildfire occurrence due to global climate changes urged the improvement of present wildfire growth prediction and evaluation methods. This study aimed to propose novel solutions to their two primary limitations, including the lack of robust fuel classification method and the low spatial resolution of wildfire growth accuracy assessment while ensuring wide applicability using open data satellite missions and software. The first objective was to create a robust two-step fuel model classification method consisted of the supervised machine learning classification of generalized land cover classes in the 1st level and their individual unsupervised classification to vegetation subtypes in the 2nd level. The second objective was creating a wildfire prediction accuracy assessment method using MODIS 250 m images, which overcome the limitations of low spatial resolution while preserving sub-daily temporal resolution. The wildfire on the Korčula island in Croatia was analyzed in the study, being specific for its long duration from 18 to 24 July 2015. The wildfire ignition occurred in the isolated area, which prolonged the response time from emergency agencies. Random Forest (RF) with input Landsat 8 spectral bands and indices resulted in the highest classification accuracy in the 1st classification level with an overall agreement of 83.6%. The vegetation subclasses from the 2nd classification level were matched to the 13 standard fuel models for the input in FARSITE software. The predicted wildfire evaluation showed the highest mean accuracy of 0.906 for the first two days, which decreased to 0.722 in the latter stages of the active wildfire caused by overprediction. The proposed two-step fuel model classification presented a cost-efficient solution to the fuel map creation in any part of the world, with a disadvantage of no in-situ ground truth identification and accuracy assessment for 2nd classification level. The evaluation of wildfire growth prediction with 250 m images enabled high spatial and temporal resolution of the assessment, while its limitations of wildfire overprediction and the negative effects of wildfire smoke in MODIS images should be addressed in future research. [Display omitted] • Wildfire growth was predicted and evaluated using Landsat and MODIS images. • The fuel model was classified using a robust two-step classification method. • Wildfire perimeters were evaluated using 250 m daily NDVI time series. • FARSITE prediction mean accuracy was 0.906 in the first two days of active wildfire. • Proposed cost-efficient methods support the decision-making in web services. [ABSTRACT FROM AUTHOR]

The precision fertilization system is the basis for upgrading conventional intensive agricultural production, while achieving both high and quality yields and minimizing the negative impacts on the environment. This research aims to present the application of both conventional and modern prediction methods in precision fertilization by integrating agronomic components with the spatial component of interpolation and machine learning. While conventional methods were a cornerstone of soil prediction in the past decades, new challenges to process larger and more complex data have reduced their viability in the present. Their disadvantages of lower prediction accuracy, lack of robustness regarding the properties of input soil sample values and requirements for extensive cost- and time-expensive soil sampling were addressed. Specific conventional (ordinary kriging, inverse distance weighted) and modern machine learning methods (random forest, support vector machine, artificial neural networks, decision trees) were evaluated according to their popularity in relevant studies indexed in the Web of Science Core Collection over the past decade. As a shift towards increased prediction accuracy and computational efficiency, an overview of state-of-the-art remote sensing methods for improving precise fertilization was completed, with the accent on open-data and global satellite missions. State-of-the-art remote sensing techniques allowed hybrid interpolation to predict the sampled data supported by remote sensing data such as high-resolution multispectral, thermal and radar satellite or unmanned aerial vehicle (UAV)-based imagery in the analyzed studies. The representative overview of conventional and modern approaches to precision fertilization was performed based on 121 samples with phosphorous pentoxide (P2O5) and potassium oxide (K2O) in a common agricultural parcel in Croatia. It visually and quantitatively confirmed the superior prediction accuracy and retained local heterogeneity of the modern approach. The research concludes that remote sensing data and methods have a significant role in improving fertilization in precision agriculture today and will be increasingly important in the future. [ABSTRACT FROM AUTHOR]

The accurate detection and treatment of weeds in agricultural fields is a necessary procedure for managing crop yield and avoiding herbicide pollution. With the emergence of unmanned aerial vehicles (UAV), the ability to acquire spatial data at the desired spatial and temporal resolution became available, and the resulting input data met high standards for weed management. In this paper, we tested four independent classification algorithms for the creation of weed maps, combining automatic and manual methods, as well as object-based and pixel-based classification approaches, which were used separately on two subsets. Input UAV data were collected using a low-cost RGB camera due to its affordability compared to multispectral cameras. Classification algorithms were based on the random forest machine learning algorithm for weed and bare soil extraction, following an unsupervised classification with the K-means algorithm for further estimation of weeds and bare soil presence in non-weed and non-soil areas. Of the four classification algorithms tested, the automatic object-based classification method achieved the highest classification accuracy, resulting in an overall accuracy of 89.0% for subset A and 87.1% for subset B. Automatic classification methods were robustly developed, using at least 0.25% of the scene size as the training data set in all circumstances anticipated for the random forest classification algorithm to operate. The use of the algorithm resulted in weed maps consisting of zoned classes and covering areas with similar biological properties, making them ready for use as inputs in weed treatments that use agricultural machinery. [ABSTRACT FROM AUTHOR]