scholarly journals A Methodological Comparison of Three Models for Gully Erosion Susceptibility Mapping in the Rural Municipality of El Faid (Morocco)

2021 ◽  
Vol 13 (2) ◽  
pp. 682
Author(s):  
Ali Azedou ◽  
Said Lahssini ◽  
Abdellatif Khattabi ◽  
Modeste Meliho ◽  
Nabil Rifai
Keyword(s):  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Area Under The Curve ◽  
Gully Erosion ◽  
Training Dataset ◽  
Slope Aspect ◽  
Topographic Wetness Index ◽  
Predictive Capacity ◽  
Slope Length ◽  
Rural Municipality

Erosion is the main threat to sustainable water and soil management in Morocco. Located in the Souss-Massa watershed, the rural municipality of El Faid remains an area where gully erosion is a major factor involved in soil degradation and flooding. The aim of this study is to predict the spatial distribution of gully erosion at the scale of this municipality and to evaluate the predictive capacity of three prediction methods (frequency ratio (FR), logistic regression (LR), and random forest (RF)) for the characterization of gullying vulnerability. Twelve predisposing factors underlying gully formation were considered and mapped (elevation, slope, aspect, plane curvature, slope length (SL), stream power index (SPI), composite topographic index (CTI), land use, topographic wetness index (TWI), normalized difference vegetation index (NDVI), lithology, and vegetation cover (C factor). Furthermore, 894 gullies were digitized using high-resolution imagery. Seventy-five percent of the gullies were randomly selected and used as a training dataset, whereas the remaining 25% were used for validation purposes. The prediction accuracy was evaluated using area under the curve (AUC). Results showed that the factor that most contributed to the prevalence of gullying was topographic (slope, CTI, LS). Furthermore, the fitted models revealed that the RF model had a better prediction quality, with the best AUC (91.49%). The produced maps represent a valuable tool for sustainable management, land conservation, and protecting human lives against natural hazards (floods).

scholarly journals A Semi-Automated Object-Based Gully Networks Detection Using Different Machine Learning Models: A Case Study of Bowen Catchment, Queensland, Australia

Sensors ◽  
2019 ◽  
Vol 19 (22) ◽  
pp. 4893 ◽  
Author(s):  
Hejar Shahabi ◽  
Ben Jarihani ◽  
Sepideh Tavakkoli Piralilou ◽  
David Chittleborough ◽  
Mohammadtaghi Avand ◽  
...  
Keyword(s):  
Machine Learning ◽  
Image Segmentation ◽  
Vegetation Index ◽  
Scale Parameter ◽  
Slope Aspect ◽  
Support Vector ◽  
Topographic Wetness Index ◽  
Slope Length ◽  
Optimal Scale ◽  
Object Based

Gully erosion is a dominant source of sediment and particulates to the Great Barrier Reef (GBR) World Heritage area. We selected the Bowen catchment, a tributary of the Burdekin Basin, as our area of study; the region is associated with a high density of gully networks. We aimed to use a semi-automated object-based gully networks detection process using a combination of multi-source and multi-scale remote sensing and ground-based data. An advanced approach was employed by integrating geographic object-based image analysis (GEOBIA) with current machine learning (ML) models. These included artificial neural networks (ANN), support vector machines (SVM), and random forests (RF), and an ensemble ML model of stacking to deal with the spatial scaling problem in gully networks detection. Spectral indices such as the normalized difference vegetation index (NDVI) and topographic conditioning factors, such as elevation, slope, aspect, topographic wetness index (TWI), slope length (SL), and curvature, were generated from Sentinel 2A images and the ALOS 12-m digital elevation model (DEM), respectively. For image segmentation, the ESP2 tool was used to obtain three optimal scale factors. On using object pureness index (OPI), object matching index (OMI), and object fitness index (OFI), the accuracy of each scale in image segmentation was evaluated. The scale parameter of 45 with OFI of 0.94, which is a combination of OPI and OMI indices, proved to be the optimal scale parameter for image segmentation. Furthermore, segmented objects based on scale 45 were overlaid with 70% and 30% of a prepared gully inventory map to select the ML models’ training and testing objects, respectively. The quantitative accuracy assessment methods of Precision, Recall, and an F1 measure were used to evaluate the model’s performance. Integration of GEOBIA with the stacking model using a scale of 45 resulted in the highest accuracy in detection of gully networks with an F1 measure value of 0.89. Here, we conclude that the adoption of optimal scale object definition in the GEOBIA and application of the ensemble stacking of ML models resulted in higher accuracy in the detection of gully networks.

scholarly journals Spatial Modelling of Gully Erosion Using GIS and R Programing: A Comparison among Three Data Mining Algorithms

2018 ◽  
Vol 8 (8) ◽  
pp. 1369 ◽  
Author(s):  
Alireza Arabameri ◽  
Biswajeet Pradhan ◽  
Hamid Reza Pourghasemi ◽  
Khalil Rezaei ◽  
Norman Kerle
Keyword(s):  
Data Mining ◽  
Spatial Relationship ◽  
Area Under The Curve ◽  
Regression Tree ◽  
Drainage Density ◽  
Gully Erosion ◽  
Slope Aspect ◽  
Topographic Wetness Index ◽  
Boosted Regression Tree ◽  
Area Index

Gully erosion triggers land degradation and restricts the use of land. This study assesses the spatial relationship between gully erosion (GE) and geo-environmental variables (GEVs) using Weights-of-Evidence (WoE) Bayes theory, and then applies three data mining methods—Random Forest (RF), boosted regression tree (BRT), and multivariate adaptive regression spline (MARS)—for gully erosion susceptibility mapping (GESM) in the Shahroud watershed, Iran. Gully locations were identified by extensive field surveys, and a total of 172 GE locations were mapped. Twelve gully-related GEVs: Elevation, slope degree, slope aspect, plan curvature, convergence index, topographic wetness index (TWI), lithology, land use/land cover (LU/LC), distance from rivers, distance from roads, drainage density, and NDVI were selected to model GE. The results of variables importance by RF and BRT models indicated that distance from road, elevation, and lithology had the highest effect on GE occurrence. The area under the curve (AUC) and seed cell area index (SCAI) methods were used to validate the three GE maps. The results showed that AUC for the three models varies from 0.911 to 0.927, whereas the RF model had a prediction accuracy of 0.927 as per SCAI values, when compared to the other models. The findings will be of help for planning and developing the studied region.

GIS-based automated mapping hazardous exogenous processes in the Tunka depression

2020 ◽  
Vol 957 (3) ◽  
pp. 8-20
Author(s):  
O.V. Bezgodova ◽  
Е.А. Rasputina
Keyword(s):  
Spatial Analysis ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Digital Terrain Model ◽  
Model Space ◽  
Slope Aspect ◽  
Rock Falls ◽  
Slope Length ◽  
Space Images ◽  
Automated Mapping

The authors consider automated mapping and spatial analysis of hazardous exogenous processes using a digital elevation model, space images and GIS technologies. A highly detailed map of dangerous exogenous processes within the Tunka depression has been created. At the initial stage, using digital terrain model data, the maps of natural characteristics, which cause the distribution of exogenous processes, were made. This is an elevation map, steepness of slopes and slope aspect maps, normalized difference vegetation index map (NDVI), topographic humidity index and slope length (LS) factor maps. Additional information was obtained from the analysis of space images, landscape conditions and field data. Certain sets and ranges of indicators that characterize the spatial distribution of each class of hazardous exogenous processes were chosen. Seven classes of the most intense hazardous exogenous processes were revealed. The main hazardous exogenous processes in the territory of the Tunka depression were noted on a vector basis using the method of spatial analysis in geographic information systems. Fluvial, slope erosion, and cryogenic-slope classes are most common. Fluvial class of exogenous processes is most dangerous due to mudflows within the valleys of small and temporary watercourses. Slope erosion class is represented by gully erosion, and cryogenic-slope is most often manifested in the form of solifluction. The local danger is represented by screes and rock falls, which are included in the group of gravitational-slope processes.

scholarly journals Spatial analysis of damaged vegetation in the Mianyuan River basin after the Wenchuan Earthquake

2015 ◽  
Vol 3 (5) ◽  
pp. 3225-3250
Author(s):  
H. Z. Zhang ◽  
J. R. Fan ◽  
X. M. Wang ◽  
T. H. Chi ◽  
L. Peng
Keyword(s):  
Slope Stability ◽  
Wenchuan Earthquake ◽  
River Basin ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Probabilistic Approach ◽  
Model Development ◽  
Slope Aspect ◽  
2008 Wenchuan Earthquake ◽  
Elevation Model

Abstract. The 2008 Wenchuan earthquake destroyed large areas of vegetation. Presently, these areas of damaged vegetation are at various stages of recovery. In this study, we present a probabilistic approach for slope stability analysis that quantitatively relates data on earthquake-damaged vegetation with slope stability in a given river basin. The Mianyuan River basin was selected for model development, and earthquake-damaged vegetation and post-earthquake recovery conditions were identified via the normalized difference vegetation index (NDVI), from multi-temporal (2001–2014) remote sensing images. DSAL (digital elevation model, slope, aspect, and lithology) spatial zonation was applied to characterize the survival environments of vegetation, which were used to discern the relationships between successful vegetation regrowth and environmental conditions. Finally, the slope stability susceptibility model was trained through multivariate analysis of earthquake-damaged vegetation and its controlling factors (i.e. topographic environments and material properties). Application to the Subao River basin validated the proposed model, showing that most of the damaged vegetation areas have high susceptibility levels (88.1% > susceptibility level 3, and 61.5% > level 4). Our modelling approach may also be valuable for use in other regions prone to landslide hazards.

scholarly journals Identification of Surface Water Storing Sites Using Topographic Wetness Index (TWI) and Normalized Difference Vegetation Index (NDVI)

2018 ◽  
Vol 8 ◽  
pp. 91-100
Author(s):  
Belete Berhanu ◽  
Ethiopia Bisrat
Keyword(s):  
Surface Water ◽  
Water Resource ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Cost Effective ◽  
Water Shortage ◽  
Sample Volume ◽  
Runoff Generation ◽  
Topographic Wetness Index ◽  
Wetness Index

Ethiopia is endowed with water and has a high runoff generation area compared to many countries, but the total stored water only goes up to approximately 36BCM. The problem of water shortage in Ethiopia emanates from the seasonality of rainfall and the lack of infrastructure for storage to capture excess runoff during flood seasons. Based on this premise, a method for a syndicate use of topography, land use and vegetation was applied to locate potential surface water storing sites. The steady-state Topographic Wetness Index (TWI) was used to represent the spatial distribution of water flow and water stagnating across the study area and the Normalized Difference Vegetation Index (NDVI) was used to detect surface water through multispectral analysis. With this approach, a number of water storing sites were identified in three categories: primary sources (water bodies based), secondary sources (Swampy/wetland based) and tertiary sources (the land based). A sample volume analysis for the 120354 water storing sites in category two, gives a 44.92BCM potential storing capacity with average depth of 4 m that improves the annual storage capacity of the country to 81BCM (8.6 % of annual renewable water sources). Finally, the research confirmed the TWI and NDVI based approach for water storing sites works without huge and complicated earth work; it is cost effective and has the potential of solving complex water resource challenges through spatial representation of water resource systems. Furthermore, the application of remote sensing captures temporal diversity and includes repetitive archives of data, enabling the monitoring of areas, even those that are inaccessible, at regular intervals.

scholarly journals Mapping Forest Fire Risk—A Case Study in Galicia (Spain)

2020 ◽  
Vol 12 (22) ◽  
pp. 3705
Author(s):  
Ana Novo ◽  
Noelia Fariñas-Álvarez ◽  
Joaquín Martínez-Sánchez ◽  
Higinio González-Jorge ◽  
José María Fernández-Alonso ◽  
...  
Keyword(s):  
Forest Fire ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Fire Risk ◽  
Fire Regimes ◽  
Slope Aspect ◽  
Fire Weather Index ◽  
Novel Approach ◽  
Fuel Model ◽  
Forest Fire Risk

The optimization of forest management in roadsides is a necessary task in terms of wildfire prevention in order to mitigate their effects. Forest fire risk assessment identifies high-risk locations, while providing a decision-making support about vegetation management for firefighting. In this study, nine relevant parameters: elevation, slope, aspect, road distance, settlement distance, fuel model types, normalized difference vegetation index (NDVI), fire weather index (FWI), and historical fire regimes, were considered as indicators of the likelihood of a forest fire occurrence. The parameters were grouped in five categories: topography, vegetation, FWI, historical fire regimes, and anthropogenic issues. This paper presents a novel approach to forest fire risk mapping the classification of vegetation in fuel model types based on the analysis of light detection and ranging (LiDAR) was incorporated. The criteria weights that lead to fire risk were computed by the analytic hierarchy process (AHP) and applied to two datasets located in NW Spain. Results show that approximately 50% of the study area A and 65% of the study area B are characterized as a 3-moderate fire risk zone. The methodology presented in this study will allow road managers to determine appropriate vegetation measures with regards to fire risk. The automation of this methodology is transferable to other regions for forest prevention planning and fire mitigation.

scholarly journals Predicting the distribution of forest tree species using topographic variables and vegetation index in eastern Acre, Brazil

2015 ◽  
Vol 45 (2) ◽  
pp. 167-174 ◽  
Author(s):  
Symone Maria de Melo FIGUEIREDO ◽  
Eduardo Martins VENTICINQUE ◽  
Evandro Orfanó FIGUEIREDO ◽  
Evandro José Linhares FERREIRA
Keyword(s):  
Tree Species ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Elevation Gradient ◽  
Forest Tree ◽  
Entropy Method ◽  
Slope Aspect ◽  
Tropical Species ◽  
Ecological Requirements ◽  
Timber Tree

Species distribution modeling has relevant implications for the studies of biodiversity, decision making about conservation and knowledge about ecological requirements of the species. The aim of this study was to evaluate if the use of forest inventories can improve the estimation of occurrence probability, identify the limits of the potential distribution and habitat preference of a group of timber tree species. The environmental predictor variables were: elevation, slope, aspect, normalized difference vegetation index (NDVI) and height above the nearest drainage (HAND). To estimate the distribution of species we used the maximum entropy method (Maxent). In comparison with a random distribution, using topographic variables and vegetation index as features, the Maxent method predicted with an average accuracy of 86% the geographical distribution of studied species. The altitude and NDVI were the most important variables. There were limitations to the interpolation of the models for non-sampled locations and that are outside of the elevation gradient associated with the occurrence data in approximately 7% of the basin area. Ceiba pentandra (samaúma), Castilla ulei (caucho) and Hura crepitans (assacu) is more likely to occur in nearby water course areas. Clarisia racemosa (guariúba), Amburana acreana (cerejeira), Aspidosperma macrocarpon (pereiro), Apuleia leiocarpa (cumaru cetim), Aspidosperma parvifolium (amarelão) and Astronium lecointei (aroeira) can also occur in upland forest and well drained soils. This modeling approach has potential for application on other tropical species still less studied, especially those that are under pressure from logging.

scholarly journals Wildfire Probability Mapping: Bivariate vs. Multivariate Statistics

2019 ◽  
Vol 11 (6) ◽  
pp. 618 ◽  
Author(s):  
Abolfazl Jaafari ◽  
Davood Mafi-Gholami ◽  
Binh Thai Pham ◽  
Dieu Tien Bui
Keyword(s):  
Multivariate Statistics ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Predictive Accuracy ◽  
Rank Test ◽  
Wind Effect ◽  
Ensemble Modeling ◽  
Residential Areas ◽  
Topographic Wetness Index ◽  
Distribution Maps

Wildfires are one of the most common natural hazards worldwide. Here, we compared the capability of bivariate and multivariate models for the prediction of spatially explicit wildfire probability across a fire-prone landscape in the Zagros ecoregion, Iran. Dempster–Shafer-based evidential belief function (EBF) and the multivariate logistic regression (LR) were applied to a spatial dataset that represents 132 fire events from the period of 2007–2014 and twelve explanatory variables (altitude, aspect, slope degree, topographic wetness index (TWI), annual temperature, and rainfall, wind effect, land use, normalized difference vegetation index (NDVI), and distance to roads, rivers, and residential areas). While the EBF model successfully characterized each variable class by four probability mass functions in terms of wildfire probabilities, the LR model identified the variables that have a major impact on the probability of fire occurrence. Two distribution maps of wildfire probability were developed based upon the results of each model. In an ensemble modeling perspective, we combined the two probability maps. The results were verified and compared by the receiver operating characteristic (ROC) and the Wilcoxon Signed-Rank Test. The results showed that although an improved predictive accuracy (AUC = 0.864) can be achieved via an ensemble modeling of bivariate and multivariate statistics, the models fail to individually provide a satisfactory prediction of wildfire probability (EBFAUC = 0.701; LRAUC = 0.728). From these results, we recommend the employment of ensemble modeling approaches for different wildfire-prone landscapes.

scholarly journals Multi-scale selection models predict breeding habitat for two Arctic-breeding raptor species

2020 ◽  
Vol 6 (1) ◽  
pp. 24-40
Author(s):  
Philippe Galipeau ◽  
Alastair Franke ◽  
Mathieu Leblond ◽  
Joel Bêty
Keyword(s):  
Habitat Selection ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Spatial Scales ◽  
Environmental Indicators ◽  
Breeding Habitat ◽  
Slope Aspect ◽  
Peregrine Falcon ◽  
Apex Predators ◽  
Selection Models

Raptors are important environmental indicators because they are apex predators and can be sensitive to disturbance. Few studies have addressed habitat preferences of tundra-nesting raptors, and those that exist have focused on fine-scale characteristics. With increasing economic development predicted to occur throughout the Canadian Arctic, the investigation of raptor breeding habitat at broad spatial scales is required. We modeled breeding habitat selection for two raptor species on north Baffin Island, NU, Canada. During aerial surveys conducted over six breeding seasons, we documented 172 peregrine falcon (Falco peregrinus tundrius) and 160 rough-legged hawk (Buteo lagopus) nesting sites. We used these locations in conjunction with remote sensing data to build habitat selection models at three spatial scales. Topography, distance to water, and normalized difference vegetation index explained selection at all scales; slope aspect was also important at the finest scale. To validate landscape scale models, we conducted a validation survey that resulted in the detection of 45 new nests (peregrine falcon n = 21, rough-legged hawk n = 24). We did not detect any new nests in areas where model-predicted occurrence was expected to be low. Conversely, we found more than half of previously undetected nests in areas where model-predicted occurrence was expected to be high.

scholarly journals Landslide Susceptibility Mapping Using Machine Learning Algorithms and Remote Sensing Data in a Tropical Environment

2020 ◽  
Vol 17 (14) ◽  
pp. 4933 ◽  
Author(s):  
Viet-Ha Nhu ◽  
Ayub Mohammadi ◽  
Himan Shahabi ◽  
Baharin Bin Ahmad ◽  
Nadhir Al-Ansari ◽  
...  
Keyword(s):  
Landslide Susceptibility ◽  
Predictive Value ◽  
Vegetation Index ◽  
Performance Metrics ◽  
Characteristic Curve ◽  
Google Earth ◽  
Machine Learning Algorithms ◽  
Landslide Susceptibility Mapping ◽  
Slope Aspect ◽  
Topographic Wetness Index

We used AdaBoost (AB), alternating decision tree (ADTree), and their combination as an ensemble model (AB-ADTree) to spatially predict landslides in the Cameron Highlands, Malaysia. The models were trained with a database of 152 landslides compiled using Synthetic Aperture Radar Interferometry, Google Earth images, and field surveys, and 17 conditioning factors (slope, aspect, elevation, distance to road, distance to river, proximity to fault, road density, river density, normalized difference vegetation index, rainfall, land cover, lithology, soil types, curvature, profile curvature, stream power index, and topographic wetness index). We carried out the validation process using the area under the receiver operating characteristic curve (AUC) and several parametric and non-parametric performance metrics, including positive predictive value, negative predictive value, sensitivity, specificity, accuracy, root mean square error, and the Friedman and Wilcoxon sign rank tests. The AB model (AUC = 0.96) performed better than the ensemble AB-ADTree model (AUC = 0.94) and successfully outperformed the ADTree model (AUC = 0.59) in predicting landslide susceptibility. Our findings provide insights into the development of more efficient and accurate landslide predictive models that can be used by decision makers and land-use managers to mitigate landslide hazards.

Sign in / Sign up

Export Citation Format

Share Document