Search

The problem of mapping regions with socially-derived boundaries has been a topic of discussion in the GIS literature for many years. Fuzzy approaches have frequently been suggested as solutions, but none have been adopted. This is likely due to difficulties associated with determining suitable membership functions, which are often as arbitrary as the crisp boundaries that they seek to replace. This paper presents a novel approach to fuzzy geographical modelling that replaces the membership function with a possibility distribution that is estimated using Bayesian inference. In this method, data from multiple sources are combined to estimate the degree to which a given location is a member of a given set and the level of uncertainty associated with that estimate. The Fuzzy Bayesian Inference approach is demonstrated through a case study in which census data are combined with perceptual and behavioural evidence to model the territory of two segregated groups (Catholics and Protestants) in Belfast, Northern Ireland, UK. This novel method provides a robust empirical basis for the use of fuzzy models in GIS, and therefore has applications for mapping a range of socially-derived and otherwise vague boundaries.

The problem of mapping regions with socially-derived boundaries has been a topic of discussion in the GIS literature for many years. Fuzzy approaches have frequently been suggested as solutions, but none have been adopted. This is likely due to difficulties associated with determining suitable membership functions, which are often as arbitrary as the crisp boundaries that they seek to replace. This paper presents a novel approach to fuzzy geographical modelling that replaces the membership function with a possibility distribution that is estimated using Bayesian inference. In this method, data from multiple sources are combined to estimate the degree to which a given location is a member of a given set and the level of uncertainty associated with that estimate. The Fuzzy Bayesian Inference approach is demonstrated through a case study in which census data are combined with perceptual and behavioural evidence to model the territory of two segregated groups (Catholics and Protestants) in Belfast, Northern Ireland, UK. This novel method provides a robust empirical basis for the use of fuzzy models in GIS, and therefore has applications for mapping a range of socially-derived and otherwise vague boundaries.

Water harvesting has a long history, but still plays an important role today by increasing crop productivity, combatting erosion, and improving water supplies. Geographical Information Systems (GIS) are used extensively to assess the suitability of sites for water harvesting but available tools fail to consider the synoptic topography of sites. Here, we report the creation of a novel, automated tool – “SiteFinder” – that evaluates potential locations by automatically calculating site-specific information, including structure parameters (height, length, and volume) and descriptors of the zone affected by the structure (storage capacity and area of influence) and the catchment area. Innovatively, compared to existing tools of this kind, SiteFinder works within a GIS environment. Thus, it allows the possibility of combining its outputs with larger Multi-Criteria Decision-Making processes to consider other bio-physical, socio-economic, and environmental factors. It utilises a Digital Elevation Model (DEM) and automatically analyses thousands of potential sites, computing site characteristics for different barrier heights that are dependent on the surrounding topography. It outputs values of eight parameters to aid planners in assessing the characteristics of sites as to their suitability for water harvesting. We conducted case studies using 30 × 30 m gridded DEMs to automatically evaluate several thousand sites and, by filtering the tool outputs, successfully identified sites with characteristics appropriate for scenarios at three spatial scales: large dams for nationally significant water supply reservoirs (383 sites analysed; 5 filtered sites with barriers up to 30 m in height); large gully erosion control dams for regional-scale interventions (4,586 sites analysed; 6 filtered sites with barriers up to 3.6 m in height); and local, community-based earth embankment projects (801 sites analysed; 6 filtered sites with barriers up to 2 m in height). A higher resolution (1 × 1 m) terrain elevation model, derived from open-source airborne survey data, was used to assess the veracity of these results. Correlations between the barrier length, impounded area and storage volume capacity derived from the two different resolution data sets were all strongly significant (Spearman’s rank correlation, p < 0.001); and normalised root mean square errors were 9%, 15% and 16% for these parameters, respectively.

A novel procedure is presented for identifying water harvesting sites and their potential storage capacities from spatial data of water scarce regions. Via analysis of Digital Terrain Models, the procedure identifies appropriate sites, applies barriers (representing dams etc.) and combines these with topographic contours to create polygons representing areas impounded by the barriers. These polygons are then intersected with elevation data to obtain the potential storage volume created by each barrier. Vegetation cover within each polygon is also extracted, from spectral-based rasters. This improves on existing water harvesting site selection techniques which consider parameters (e.g. slope) on a cell-by- cell basis.

This work investigates the potential of combining the outputs of multiple low-cost sensor technologies for the direct measurement of spatio-temporal variations in phenomena that exist at the interface between our bodies and the environment. The example used herein is the measurement of personal exposure to traffic pollution, which may be considered as a function of the concentration of pollutants in the air, and the frequency and volume of that air which enters our lungs. The sensor-based approach described in this paper removes the ‘traditional’ requirements to either model or interpolate pollution levels, or to make assumptions about the physiology of an individual. Rather, a wholly empirical analysis into pollution exposure is possible, based upon high-resolution spatio-temporal data drawn from sensors for NO2, nasal airflow, and location (GPS). Data are collected via a custom smartphone application and mapped to give an unprecedented insight into exposure to traffic pollution at the individual level. Whilst the quality of data from low-cost miniaturised sensors is not suitable for all applications, there certainly are many applications for which these data would be well suited, particularly those in the field of citizen science. This paper demonstrates both the potential and limitations of sensor-based approaches, and discusses the wider relevance of these technologies for the advancement of citizen science.

The Dubračina catchment in Croatia experiences high levels of sediment erosion due to a combination of geological and topographical factors combined with high levels or rainfall and varied land cover characteristics. This paper assesses the sensitivity of the Gavrilović Erosion Potential Method to different land cover/use inputs with different spatial characteristics. We conclude that volume and spatial distribution of eroded material is best predicted when land cover derived from Landsat 8 imagery is utilised.

In recent decades there has been a significant development of erosion assessment methods that simultaneously followed the development of computer technologies, as well as Geographic Information Systems (GIS) and Satellite Imagery, thus enabling more detailed information`s about topography, land use and vegetation cover, as well as broaden the possibilities for the application of more demanding erosion analysis. There are several papers that deal with the application of various erosion assessment methods depending on the needed scale [1, 2, 3] (from global to catchment size), erosion type (gully, rill, bank, sheet) [3] and their assessments by criteria such as prediction accuracy, erosion processes, needed data and calibration [8]. Within this paper twenty-two different erosion assessment methods are analysed and compared with the purpose to define the relevance of each used criterion, better understanding of erosion processes, as well as to give future guidance for simplifying the procedure of choosing the appropriate method based on available data and relevant criteria. Methods encompassed with this analysis are Pacific Southwest Inter-Agency Committee (PSIAC), PSIAC adapted version, The vegetation-surface material-drainage density (VSD), Erosion Potential Method–Gavrilovic (EPM), Factorial Scouring Model (FSM), Erosion hazard units (EHU), Soil Loss Estimation Model for Southern Africa (SLEMSA), CORINE erosion risk maps, Coleman and Scatena scoring model (CSSM), Fleming and Kadhimi scoring model (FKSM), Wallingford scoring model (WSM), Universal Soil Loss Equation (USLE), Revised Universal Soil Loss Equation (RUSLE), RIVM Model, INRA Model, SCALES Model, Fournier, Water Erosion Prediction Model (WEPP), Soil and Water Assessment Tool (SWAT), Morgan-Morgan-Finney (MMF), Annualized Agricultural Non-Point Source Pollution (AGNPS) and Modified Universal Soil Loss Equation (MUSLE) [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]. There are fifty one used criteria within these methods that can be divided into ten main groups (soil, climate and precipitation, runoff, water network, topography, vegetation cover and land use, upland erosion, channel erosion and sediment transport, catchment characteristics and other). Ten most used criteria are precipitation, erosivity or rain intensity (72.73%), slope angle (72.73%), soil erodibility (68.18%), percentage of vegetation cover (40.91%), cover type (36.36%), runoff (31.82%), agricultural practice (31.82%), soil type (22.73%), slope length (22.73%), agricultural land (13.64%), all of which represent only five out of ten criteria groups (topography, climate and precipitation, soil, vegetation cover and land use, runoff). By group statistic, where at least one of the group criteria is used in each method, vegetation cover and land use, can be considered the most significant one, with the use percentage of 95.45%, followed both by soil and topography groups with 86.36% and climate and precipitation with 81.82%. There is a minimum gap of 45% between the use of first four group criteria and the rest of the groups (eg. Runoff is fifth by 36.36%). Taking into consideration conducted analysis and complementing it with the knowledge about erosion processes obtained from the literature, significance of the criteria can be concluded. Since, rainfall is considered the most important detaching agent and erodibility and type of the soil define susceptibility of the soil to detachment, these criteria can be considered the most important ones. When detached, soil is transported further by erosion agents (eg. running water) during which topography (eg. slope angle) has a major impact upon the distance, speed and pathways for the runoff and sediment transport, imposing this criteria as relevant when making methodology selection. Agricultural practice, the growth cycle of the plants, % of vegetation cover, the constructions sites, excavation of mineral resources, form vegetation cover and land use group. If not managed properly, this criteria can contribute to the increase of erosion detached sediment, and therefore needs to be taken into consideration [12, 13, 14]. The first step to predict erosion and its severity on the area of interest is choosing the methodology to apply. The restrictions of scale applicability of a method, and type of erosion the method deals with, has already been covered within literature [1, 2, 3]. The accessibility of a data is often the crucial factor in the process of method selection. For this reason it is necessary to take into consideration all four criteria (scale, erosion type, criteria significance and available data) in order to make the best method selection. Today, continuous development of technology has enabled new and more detailed spatially variant data, as well as provided various different sources for the same data category (such as landcover). That also brought the challenges for appropriate data selection within various sources, where each decision made will have its own impact on the final result. The proposed methodology has been applied on Dubracina Catchment Area in Croatia by which the choice of applicable methods was narrowed to: Gavrilovic, FKSM and Fournier method. The Gavrilovic method has been developed for catchments with karstic terrain and torrential rivers, as well as taking into consideration the previously mentioned significant criteria, all of which are available and correspond to the Dubracina Catchment. The challenge was to choose one over several available data of the same type but different source, or combining them into one secondary data with the help of GIS (eg. vegetation cover and land use). Also some elements included in the method were not available for previous researches in such detail and spatial variance (eg. drainage density) and some were available as limited number of point sources (eg. precipitation) thus making additional challenge in transferring them to spatially variant data.

Smartphone applications are driving a growing interest in 3D modelling – in particular the use of LiDAR for modelling city landscapes at high levels of precision and accuracy. In such dynamic environments, where we wish to make decisions based on places of interest that are in the (rapidly changing) field of view, it is critical that we address performance issues in visibility analysis. This paperexplores optimisation of a point-to-point line-of-sight algorithm, outlining a variety of line of sight sampling strategies, together with a number of trials that enabled us to optimise sampling in the context of urban visibility analysis.