Potential of Sentinel-2A Data to Model Surface and Canopy Fuel Characteristics in Relation to Crown Fire Hazard

Background: Crown fires are often intense and fast spreading and hence can have serious impacts on soil, vegetation, and wildlife habitats. Fire managers try to prevent the initiation and spread of crown fires in forested landscapes through fuel management. The minimum fuel conditions necessary to initiate and propagate crown fires are known to be strongly influenced by four stand structural variables: surface fuel load (SFL), fuel strata gap (FSG), canopy base height (CBH), and canopy bulk density (CBD). However, there is often a lack of quantitative data about these variables, especially at the landscape scale. Methods: In this study, data from 123 sample plots established in pure, even-aged, Pinus radiata and Pinus pinaster stands in northwest Spain were analyzed. In each plot, an intensive field inventory was used to characterize surface and canopy fuels load and structure, and to estimate SFL, FSG, CBH, and CBD. Equations relating these variables to Sentinel-2A (S-2A) bands and vegetation indices were obtained using two non-parametric techniques: Random Forest (RF) and Multivariate Adaptive Regression Splines (MARS). Results: According to the goodness-of-fit statistics, RF models provided the most accurate estimates, explaining more than 12%, 37%, 47%, and 31% of the observed variability in SFL, FSG, CBH, and CBD, respectively. To evaluate the performance of the four equations considered, the observed and estimated values of the four fuel variables were used separately to predict the potential type of wildfire (surface fire, passive crown fire, or active crown fire) for each plot, considering three different burning conditions (low, moderate, and extreme). The results of the confusion matrix indicated that 79.8% of the surface fires and 93.1% of the active crown fires were correctly classified; meanwhile, the highest rate of misclassification was observed for passive crown fire, with 75.6% of the samples correctly classified. Conclusions: The results highlight that the combination of medium resolution imagery and machine learning techniques may add valuable information about surface and canopy fuel variables at large scales, whereby crown fire potential and the potential type of wildfire can be classified.

Download Full-text

Estimating Fuel Loads and Structural Characteristics of Shrub Communities by Using Terrestrial Laser Scanning

Remote Sensing ◽

10.3390/rs12223704 ◽

2020 ◽

Vol 12 (22) ◽

pp. 3704

Author(s):

Cecilia Alonso-Rego ◽

Stéfano Arellano-Pérez ◽

Carlos Cabo ◽

Celestino Ordoñez ◽

Juan Gabriel Álvarez-González ◽

...

Keyword(s):

Laser Scanning ◽

Goodness Of Fit ◽

Fire Hazard ◽

Structural Characteristics ◽

Fire Behavior ◽

Terrestrial Laser Scanning ◽

Fuel Load ◽

Fuel Loads ◽

Litter Depth ◽

Species Specific

Forest fuel loads and structural characteristics strongly affect fire behavior, regulating the rate of spread, fireline intensity, and flame length. Accurate fuel characterization, including disaggregation of the fuel load by size classes, is therefore essential to obtain reliable predictions from fire behavior simulators and to support decision-making in fuel management and fire hazard prediction. A total of 55 sample plots of four of the main non-tree covered shrub communities in NW Spain were non-destructively sampled to estimate litter depth and shrub cover and height for species. Fuel loads were estimated from species-specific equations. Moreover, a single terrestrial laser scanning (TLS) scan was collected in each sample plot and features related to the vertical and horizontal distribution of the cloud points were calculated. Two alternative approaches for estimating size-disaggregated fuel loads and live/dead fractions from TLS data were compared: (i) a two-steps indirect estimation approach (IE) based on fitting three equations to estimate shrub height and cover and litter depth from TLS data and then use those estimates as inputs of the existing species-specific fuel load equations by size fractions based on these three variables; and (ii) a direct estimation approach (DE), consisting of fitting seven equations, one for each fuel fraction, to relate the fuel load estimates to TLS data. Overall, the direct approach produced more balanced goodness-of-fit statistics for the seven fractions considered jointly, suggesting that it performed better than the indirect approach, with equations explaining more than 80% of the observed variability for all species and fractions, except the litter loads.

Download Full-text

Canopy bulk density and canopy base height equations for assessing crown fire hazard in Pinus radiata plantations

Canadian Journal of Forest Research ◽

10.1139/x10-237 ◽

2011 ◽

Vol 41 (4) ◽

pp. 839-850 ◽

Cited By ~ 16

Author(s):

Ana Daría Ruiz-González ◽

Juan Gabriel Álvarez-González

Keyword(s):

Bulk Density ◽

Pinus Radiata ◽

Fire Hazard ◽

Direct Action ◽

Quantitative Information ◽

Wildlife Habitat ◽

Vertical Profiles ◽

Crown Fire ◽

Crown Fires ◽

Canopy Bulk Density

Crown fires combine high rates of spread, flame lengths, and intensities, making it virtually impossible to control them by direct action and having significant impact on soils, vegetation, and wildlife habitat. For these reasons, fire managers have great interest in preventive silviculture of forested landscapes to avoid the initiation and propagation of crown fires. The minimum conditions necessary to initiate and propagate crown fires are assumed to be strongly influenced by the stand structural variables canopy bulk density (CBD) and canopy base height (CBH). However, there is a lack of quantitative information on these variables and how to estimate them. To characterize the aerial fuel layers of Pinus radiata D. Don, the vertical profiles of canopy fuel in 180 sample plots of pure and even-aged P. radiata plantations were analysed. Effective CBD and CBH were obtained from the vertical profiles, and equations relating these variables to common stand variables were fitted simultaneously. Inclusion of the fitted equations in existing dynamic growth models, together with the use of current fire behaviour and hazard prediction tools, will provide a decision support system for assessing the crown fire potential of different silvicultural alternatives for this species.

Download Full-text

Estimación de la distribución vertical de combustibles finos del dosel de copas en masas de Pinus sylvestris empleando datos LiDAR de baja densidad

Revista de Teledetección ◽

10.4995/raet.2019.11241 ◽

2019 ◽

pp. 1 ◽

Cited By ~ 1

Author(s):

L. A. Fidalgo-González ◽

S. Arellano-Pérez ◽

J. G. Álvarez-González ◽

F. Castedo-Dorado ◽

A. D. Ruiz-González ◽

...

Keyword(s):

Vertical Distribution ◽

Fire Hazard ◽

National Forest Inventory ◽

Fuel Load ◽

Crown Fire ◽

Cross Model ◽

Canopy Fuels ◽

Structural Variables ◽

Fire Initiation ◽

Using Data

Canopy fuel load, canopy bulk density and canopy base height are structural variables used to predict crown fire initiation and spread. Direct measurement of these variables is not functional, and they are usually estimated indirectly by modelling. Advances in fire behaviour modelling require accurate and landscape scale estimates of the complete vertical distribution of canopy fuels. The goal of the present study is to model the vertical profile of available canopy fuels in Scots pine stands by using data from the Spanish national forest inventory and low-density LiDAR data (0.5 first returns m–2) provided by Spanish PNOA project (Plan Nacional de Ortofotografía Aérea). In a first step, the vertical distribution of the canopy fuel load was modelled using the Weibull probability density function. In a second step, a system of models was fitted to relate the canopy variables to Lidar-derived metrics. Models were fitted simultaneously to compensate the effects of the inherent cross-model correlation between errors. Heteroscedasticity was also analyzed, but correction in the fitting process was not necessary. The estimated canopy fuel load profiles from LiDAR-derived metrics explained 41% of the variation in canopy fuel load in the analysed plots. The proposed models can be used to assess the effectiveness of different forest management alternatives for reducing crown fire hazard.

Download Full-text

Thinning Effect on Fuel Load and Crown Fire Hazard

Crisis and Emergency Management Theory and Praxis ◽

10.14251/crisisonomy.2016.12.8.121 ◽

2016 ◽

Vol 12 (8) ◽

pp. 121-130

Author(s):

Byung Doo Lee ◽

◽

Yeong Tae Bae ◽

Sung Cheol Jung ◽

◽

...

Keyword(s):

Fire Hazard ◽

Fuel Load ◽

Crown Fire ◽

Thinning Effect

Download Full-text

Predictive Modelling of Employee Turnover in Indian IT Industry Using Machine Learning Techniques

Vision The Journal of Business Perspective ◽

10.1177/0972262918821221 ◽

2019 ◽

Vol 23 (1) ◽

pp. 12-21 ◽

Cited By ~ 2

Author(s):

Shikha N. Khera ◽

Divya

Keyword(s):

Machine Learning ◽

Learning Algorithm ◽

Confusion Matrix ◽

Predictive Modelling ◽

Supervised Machine Learning ◽

Machine Learning Techniques ◽

Support Vector ◽

It Industry ◽

Knowledge Based ◽

Employee Attrition

Information technology (IT) industry in India has been facing a systemic issue of high attrition in the past few years, resulting in monetary and knowledge-based loses to the companies. The aim of this research is to develop a model to predict employee attrition and provide the organizations opportunities to address any issue and improve retention. Predictive model was developed based on supervised machine learning algorithm, support vector machine (SVM). Archival employee data (consisting of 22 input features) were collected from Human Resource databases of three IT companies in India, including their employment status (response variable) at the time of collection. Accuracy results from the confusion matrix for the SVM model showed that the model has an accuracy of 85 per cent. Also, results show that the model performs better in predicting who will leave the firm as compared to predicting who will not leave the company.

Download Full-text

Prediction of the critical temperature of a superconductor by using the WOA/MARS, Ridge, Lasso and Elastic-net machine learning techniques

Neural Computing and Applications ◽

10.1007/s00521-021-06304-z ◽

2021 ◽

Author(s):

Paulino José García-Nieto ◽

Esperanza García-Gonzalo ◽

José Pablo Paredes-Sánchez

Keyword(s):

Critical Temperature ◽

Optimization Algorithm ◽

Regression Models ◽

Chemical Properties ◽

Atomic Radius ◽

Elastic Net ◽

Multivariate Adaptive Regression Splines ◽

Machine Learning Techniques ◽

Humpback Whales ◽

Whale Optimization

AbstractThis study builds a predictive model capable of estimating the critical temperature of a superconductor from experimentally determined physico-chemical properties of the material (input variables): features extracted from the thermal conductivity, atomic radius, valence, electron affinity and atomic mass. This original model is built using a novel hybrid algorithm relied on the multivariate adaptive regression splines (MARS) technique in combination with a nature-inspired meta-heuristic optimization algorithm termed the whale optimization algorithm (WOA) that mimics the social behavior of humpback whales. Additionally, the Ridge, Lasso and Elastic-net regression models were fitted to the same experimental data for comparison purposes. The results of the current investigation indicate that the critical temperature of a superconductor can be successfully predicted using this proposed hybrid WOA/MARS-based model. Furthermore, the results obtained with the Ridge, Lasso and Elastic-net regression models are clearly worse than those obtained with the WOA/MARS-based model.

Download Full-text

Churn Prediction of Employees Using Machine Learning Techniques

Tehnički glasnik ◽

10.31803/tg-20210204181812 ◽

2021 ◽

Vol 15 (1) ◽

pp. 51-59

Author(s):

Nilasha Bandyopadhyay ◽

Anil Jadhav

Keyword(s):

Random Forest ◽

Confusion Matrix ◽

False Positive Rate ◽

Random Forest Classifier ◽

Working Environment ◽

Attrition Rate ◽

Machine Learning Techniques ◽

Churn Prediction ◽

Technology Industry ◽

Positive Rate

Employees are considered as the most valuable assets of any organization. Various policies have been introduced by the HR professionals to create a good working environment for them, but still, the rate of employees quitting the Technology Industry is quite high. Often the reason behind their early attrition could be due to company-related or personal issues, such as No satisfaction at the workplace, Fewer opportunities for learning, Undue Workload, Less Encouragement, and many others. This paper aims in discussing a structured way for predicting the churn rate of the employees by implementing various Classification techniques like SVM, Random Forest classifier, and Naives Bayes classifier. The performance of the classifiers was compared using metrics like Confusion Matrix, Recall, False Positive Rate, and Accuracy to determine the best model for the churn prediction. We found that among the models, the Random Forest classifier proved to be the best model for IT employee churn prediction. A Correlation Matrix was generated in the form of a heatmap to identify the important features that might impact the attrition rate.

Download Full-text

A Soft Voting Ensemble-Based Model for the Early Prediction of Idiopathic Pulmonary Fibrosis (IPF) Disease Severity in Lungs Disease Patients

Life ◽

10.3390/life11101092 ◽

2021 ◽

Vol 11 (10) ◽

pp. 1092

Author(s):

Sikandar Ali ◽

Ali Hussain ◽

Satyabrata Aich ◽

Moo Suk Park ◽

Man Pyo Chung ◽

...

Keyword(s):

Machine Learning ◽

Idiopathic Pulmonary Fibrosis ◽

Pulmonary Fibrosis ◽

Lung Disease ◽

Lung Diseases ◽

Early Stage ◽

Confusion Matrix ◽

Machine Learning Techniques ◽

Training Dataset ◽

Proposed Model

Idiopathic pulmonary fibrosis, which is one of the lung diseases, is quite rare but fatal in nature. The disease is progressive, and detection of severity takes a long time as well as being quite tedious. With the advent of intelligent machine learning techniques, and also the effectiveness of these techniques, it was possible to detect many lung diseases. So, in this paper, we have proposed a model that could be able to detect the severity of IPF at the early stage so that fatal situations can be controlled. For the development of this model, we used the IPF dataset of the Korean interstitial lung disease cohort data. First, we preprocessed the data while applying different preprocessing techniques and selected 26 highly relevant features from a total of 502 features for 2424 subjects. Second, we split the data into 80% training and 20% testing sets and applied oversampling on the training dataset. Third, we trained three state-of-the-art machine learning models and combined the results to develop a new soft voting ensemble-based model for the prediction of severity of IPF disease in patients with this chronic lung disease. Hyperparameter tuning was also performed to get the optimal performance of the model. Fourth, the performance of the proposed model was evaluated by calculating the accuracy, AUC, confusion matrix, precision, recall, and F1-score. Lastly, our proposed soft voting ensemble-based model achieved the accuracy of 0.7100, precision 0.6400, recall 0.7100, and F1-scores 0.6600. This proposed model will help the doctors, IPF patients, and physicians to diagnose the severity of the IPF disease in its early stages and assist them to take proactive measures to overcome this disease by enabling the doctors to take necessary decisions pertaining to the treatment of IPF disease.

Download Full-text

Evaluation of the Compound Topographic Index (CTI) for the location of gullies in cultivated areas of Córdoba (Spain)

10.5194/egusphere-egu21-9591 ◽

2021 ◽

Author(s):

Eduardo Luquin ◽

Yolanda Zuasti ◽

Jorge Delgado ◽

Raquel Gastesi ◽

Javier Casalí ◽

...

Keyword(s):

Goodness Of Fit ◽

Confusion Matrix ◽

Qualitative Evaluation ◽

Evaluation Tool ◽

Software Environment ◽

Topographic Index ◽

Average Value ◽

Olive Groves ◽

Compound Topographic Index ◽

And Control

The identification of areas susceptible to gully formation is an objective that has important consequences for erosion control. It allows for the optimization of resources by focusing on prevention and control efforts on the most susceptible areas, avoiding the frequent evolution of ephemeral to permanent gullies. The issue is of great interest in Spanish olive groves, many of which are affected by serious problems of gully erosion.Gullies are formed in the swales, which allows the use of topography-based tools to predict their location.The Compound Topographic Index (CTI) proposed by Thorne et al. (1986) is calculated for each pixel as an estimate of the flow capacity to cause erosion, as it includes the product of the pixel draining area and its slope. Its application requires the identification of a critical value of the CTI (CTIc), above which the potential areas of gully occurrence will be located. Using historical orthophotos, the gullies observed were digitized for 2011 in the experimental areas called Morente (11 km2 of traditional olive groves on degraded and poor vertisols) and Matasanos (6 km2 of intensive olive groves also on vertisols) and nearby area, with cereal crops.The objectives of this work are: to identify CTIc values corresponding to cultivated areas in Cordoba, mainly olive groves; to develop and evaluate an application that allows a user without great technical skills to obtain the CTI; to evaluate the capacity of this CTIc to reproduce gullies observed in nearby areas or in different time periods (2005) to establish cause-effect relationships between changes in landuse in this type of phenomenon, using the aforementioned tool.Part of the digitized gullies, representative of olive grove areas, were used to obtain the CTIc of each gully, by modifying it until the best reproduction of the gullies observed was achieved, then their average value was taken as CTIc. To calculate the CTI, a 5m resolution DEM was used, obtained from LiDAR PNOA 2014.In the framework of the Innolivar project, a desktop GIS application has been developed in a free software environment such as QGIS, which allows the calculation of the CTI. The APET tool (AGNPS Potential Ephemeral Gully Evaluation Tool) recently implemented has helped in the development of this application.The CTI calculation by the application, after the determination of the CTIc threshold, serves to identify critical areas from a DEM, which is free and available in many countries. A first qualitative evaluation by visual verification indicates a very good characterization of the gullies. Subsequently, the goodness of fit of the gully position between the digitized gullies and the app-calculated gullies according to the CTIc is evaluated quantitatively by obtaining a binary confusion matrix by lengths. In the Morente area, an error of omission of 29% and of commission of 16% was obtained.It can be concluded that the application generated that allows the application of the CTI methodology makes identification of areas susceptible to gully formation possible in an efficient and relatively simple manner, helping to achieve a more sustainable agriculture.

Download Full-text

A Multidisciplinary Approach for Evaluating Spatial and Temporal Variations in Water Quality

Water ◽

10.3390/w11040853 ◽

2019 ◽

Vol 11 (4) ◽

pp. 853 ◽

Cited By ~ 2

Author(s):

Viet Thang Le ◽

Nguyen Hong Quan ◽

Ho Huu Loc ◽

Nguyen Thi Thanh Duyen ◽

Tran Duc Dung ◽

...

Keyword(s):

Artificial Intelligence ◽

Water Quality ◽

Confusion Matrix ◽

Temporal Variations ◽

Machine Learning Techniques ◽

Spatial And Temporal Variations ◽

Support Vector ◽

Domestic Water ◽

Supply Source ◽

Artificial Intelligence Techniques

The primary goal of this study is to investigate the classification capability of several artificial intelligence techniques, including the decision tree (DT), multilayer perceptron (MLP) network, Naïve Bayes, radial basis function (RBF) network, and support vector machine (SVM) for evaluating spatial and temporal variations in water quality. The application case is the Song Quao-Ca Giang (SQ-CG) water system, a main domestic water supply source of the city of Phan Thiet in Binh Thuan province, Vietnam. To evaluate the water quality condition of the source, the government agency has initiated an extensive sampling project, collecting samples from 43 locations covering the SQ reservoir, the main canals, and the surrounding areas during 2015–2016. Different classifying models based on artificial intelligence techniques were developed to analyze the sampling data after the performances of the models were evaluated and compared using the confusion matrix, accuracy rate, and several error indexes. The results show that machine-learning techniques can be used to explicitly evaluate spatial and temporal variations in water quality.

Download Full-text