A GIS-Based Landslide Susceptibility Mapping and Variable Importance Analysis Using Artificial Intelligent Training-Based Methods

Landslides often cause significant casualties and economic losses, and therefore landslide susceptibility mapping (LSM) has become increasingly urgent and important. The potential of deep learning (DL) like convolutional neural networks (CNN) based on landslide causative factors has not been fully explored yet. The main target of this study is the investigation of a GIS-based LSM in Zanjan, Iran and to explore the most important causative factor of landslides in the case study area. Different machine learning (ML) methods have been employed and compared to select the best results in the case study area. The CNN is compared with four ML algorithms, including random forest (RF), artificial neural network (ANN), support vector machine (SVM), and logistic regression (LR). To do so, sixteen landslide causative factors have been extracted and their related spatial layers have been prepared. Then, the algorithms were trained with related landslide and non-landslide points. The results illustrate that the five ML algorithms performed suitably (precision = 82.43–85.6%, AUC = 0.934–0.967). The RF algorithm achieves the best result, while the CNN, SVM, the ANN, and the LR have the best results after RF, respectively, in this case study. Moreover, variable importance analysis results indicate that slope and topographic curvature contribute more to the prediction. The results would be beneficial to planning strategies for landslide risk management.

Estimating organic carbon in soils from Gulyantsi Municipality, Pleven District

The aim of the present study is to evaluate soil organic carbon in Gulyantsi Municipality, Pleven District. The case study area of “Ulpia Eskus” Reserve was chosen for the research. The composition of soil organic substance of Arenosols + Fluvisols is distinguished by the use of a chemical analysis and the application of the Turin method. The results show a prevalence of humic acids over fulvic acids in the sampled soil profile. Keywords: soil organic carbon, soil horizons, excavations.

Modelled exposure to motor vehicle generated noise at schools and early childhood centres.

<p><b>Motor vehicle generated noise pollution places a significant burden on the health and wellbeing of people in many urban areas and children have been identified as a particularly vulnerable group. Despite this, little is known about the extent of exposure to noise at schools and early childhood centres (ECCs), areas where children spend much of their time. To examine traffic generated noise levels at schools and ECCs, this study used the Common Noise Assessment Methods in Europe and validated the results against volunteered geographic noise measurements, using the Wellington Territorial Authority as a case study area. We examined the relationship of modelled noise values with socio-demographic variables of schools and ECCs. In addition, we assessed the relationship between modelled noise values and land use and proximity to busy roads to assess their use as proxy measures of noise. For the case study area, we found 57.7% of ECCs and 41.0% of schools exceeded the 2018 World Health Organization Environmental Noise Guidelines (53dB) and noise levels at schools and ECCs were higher compared to background levels. Schools with a higher proportion of international students, privately run ECCs, and ECCs located in the central city experienced particularly high noise levels.</b></p> <p>Compared to volunteered in situ noise measurements, our model performed reasonably: 81% of model values within 15dB of a volunteered measurement. While we found the proxy noise measurement ‘distance to busy roads’ explained 2% of the modelled noise levels in this study. Compared to proxy measures of noise, the modelled noise levels enhanced our understanding of noise level exposure. Overall, the findings of this research highlight the magnitude and inequalities of traffic generated noise pollution on children, which may be useful for guiding policy to mitigate noise pollution around schools and ECCs, such as location planning and identifying areas where ameliorating noise levels is important.</p>

Modelled exposure to motor vehicle generated noise at schools and early childhood centres.

<p><b>Motor vehicle generated noise pollution places a significant burden on the health and wellbeing of people in many urban areas and children have been identified as a particularly vulnerable group. Despite this, little is known about the extent of exposure to noise at schools and early childhood centres (ECCs), areas where children spend much of their time. To examine traffic generated noise levels at schools and ECCs, this study used the Common Noise Assessment Methods in Europe and validated the results against volunteered geographic noise measurements, using the Wellington Territorial Authority as a case study area. We examined the relationship of modelled noise values with socio-demographic variables of schools and ECCs. In addition, we assessed the relationship between modelled noise values and land use and proximity to busy roads to assess their use as proxy measures of noise. For the case study area, we found 57.7% of ECCs and 41.0% of schools exceeded the 2018 World Health Organization Environmental Noise Guidelines (53dB) and noise levels at schools and ECCs were higher compared to background levels. Schools with a higher proportion of international students, privately run ECCs, and ECCs located in the central city experienced particularly high noise levels.</b></p> <p>Compared to volunteered in situ noise measurements, our model performed reasonably: 81% of model values within 15dB of a volunteered measurement. While we found the proxy noise measurement ‘distance to busy roads’ explained 2% of the modelled noise levels in this study. Compared to proxy measures of noise, the modelled noise levels enhanced our understanding of noise level exposure. Overall, the findings of this research highlight the magnitude and inequalities of traffic generated noise pollution on children, which may be useful for guiding policy to mitigate noise pollution around schools and ECCs, such as location planning and identifying areas where ameliorating noise levels is important.</p>

Investigating the optimal integration of airborne, ship-borne, satellite and terrestrial gravity data for use in geoid determination

<p>Each gravity observation technique has different parameters and contributes to different pieces of the gravity spectrum. This means that no one gravity dataset is able to model the Earth’s gravity field completely and the best gravity map is one derived from many sources. Therefore, one of the challenges in gravity field modelling is combining multiple types of heterogeneous gravity datasets. The aim of this study is to determine the optimal method to produce a single gravity map of the Canterbury case study area, for the purposes of use in geoid modelling. This objective is realised through the identification and application of a four-step integration process: purpose, data, combination and assessment. This includes the evaluation of three integration methods: natural neighbour, ordinary kriging and least squares collocation. As geoid modelling requires the combination of gravity datasets collected at various altitudes, it is beneficial to be able to combine the dataset using an integration method which operates in a three-dimensional space. Of the three integration methods assessed, least squares collocation is the only integration method which is able to perform this type of reduction. The resulting product is a Bouguer anomaly map of the Canterbury case study area, which combines satellite altimetry, terrestrial, ship-borne, airborne, and satellite gravimetry using least squares collocation.</p>

Investigating the optimal integration of airborne, ship-borne, satellite and terrestrial gravity data for use in geoid determination

<p>Each gravity observation technique has different parameters and contributes to different pieces of the gravity spectrum. This means that no one gravity dataset is able to model the Earth’s gravity field completely and the best gravity map is one derived from many sources. Therefore, one of the challenges in gravity field modelling is combining multiple types of heterogeneous gravity datasets. The aim of this study is to determine the optimal method to produce a single gravity map of the Canterbury case study area, for the purposes of use in geoid modelling. This objective is realised through the identification and application of a four-step integration process: purpose, data, combination and assessment. This includes the evaluation of three integration methods: natural neighbour, ordinary kriging and least squares collocation. As geoid modelling requires the combination of gravity datasets collected at various altitudes, it is beneficial to be able to combine the dataset using an integration method which operates in a three-dimensional space. Of the three integration methods assessed, least squares collocation is the only integration method which is able to perform this type of reduction. The resulting product is a Bouguer anomaly map of the Canterbury case study area, which combines satellite altimetry, terrestrial, ship-borne, airborne, and satellite gravimetry using least squares collocation.</p>

Capacity of Zlatitsa Municipality (Western Bulgaria) to provide ecosystem services

The present research aims at estimating the capacity of the ecosystems in Zlatitsa Municipality to provide certain types of ecosystem services. The case study area is located in the western parts of Bulgaria, and it is a part of Sofia Province. The basis of the study is the CORINE Land Cover (CLC) classification (2018) upon which the Maes typology has been built. Fourteen (14) CLC Classes were distinguished in Zlatitsa Municipality, as well as five (5) ecosystem types. The capacity of the latter to provide ecosystem services was evaluated, based on a six-grade scale. The results of the study include maps of the provisioning, regulating, and cultural service capacity of the area, as well as an overall map of all of them. The research outcomes provided successful results, focusing on the importance of the provision of ecosystem services. They can be applied as a framework for similar studies in the neighboring municipalities.

Modeling Urban Solar Energy with High Spatiotemporal Resolution

Alternative sources of energy are being sought after in the world today, as the availability of fossil fuels and other non-renewable resources are declining. Solar energy offers a promising solution to this search as it is a less polluting renewable energy resource and can be easily converted into electricity through the usage of photovoltaic systems. This thesis focuses on the modeling of urban solar energy with high spatiotemporal resolution. A methodology was developed to estimate hourly solar PV electricity generation potential on rooftops in an urban environment using a 3-D model. A case study area of Ryerson University, Toronto was chosen and the incident solar radiation upon each building rooftop was calculated using a software tool called Ecotect Analysis 2011. Secondly, orthophotos of the case study area were digitized using Geographic Information Systems in order to eliminate undesirable rooftop objects within the model. Lastly, a software tool called HOMER was used to generate hourly solar PV electricity estimates using the values generated by the other two software tools as input parameters. It was found that hourly solar PV output followed the pattern of a binomial curve and that peak solar generation times coincided with summer peak electricity consumption hours in Ontario.

Modeling Urban Solar Energy with High Spatiotemporal Resolution

Alternative sources of energy are being sought after in the world today, as the availability of fossil fuels and other non-renewable resources are declining. Solar energy offers a promising solution to this search as it is a less polluting renewable energy resource and can be easily converted into electricity through the usage of photovoltaic systems. This thesis focuses on the modeling of urban solar energy with high spatiotemporal resolution. A methodology was developed to estimate hourly solar PV electricity generation potential on rooftops in an urban environment using a 3-D model. A case study area of Ryerson University, Toronto was chosen and the incident solar radiation upon each building rooftop was calculated using a software tool called Ecotect Analysis 2011. Secondly, orthophotos of the case study area were digitized using Geographic Information Systems in order to eliminate undesirable rooftop objects within the model. Lastly, a software tool called HOMER was used to generate hourly solar PV electricity estimates using the values generated by the other two software tools as input parameters. It was found that hourly solar PV output followed the pattern of a binomial curve and that peak solar generation times coincided with summer peak electricity consumption hours in Ontario.