Search

This study explores how combining virtual reality (VR) and augmented reality (AR) with geographic information systems (GIS) revolutionizes data visualization. It traces the historical development of these technologies and highlights key milestones that paved the way for this study's objectives. While existing platforms like Esri's software and Google Earth VR show promise, they lack complete integration for immersive GIS visualization. This gap has led to the need for a dedicated workflow to integrate selected GIS data into a game engine for visualization purposes. This study primarily utilizes QGIS for data preparation and Unreal Engine for immersive visualization. QGIS handles data management, while Unreal Engine offers advanced rendering and interactivity for immersive experiences. To tackle the challenge of handling extensive GIS datasets, this study proposes a workflow involving tiling, digital elevation model generation, and transforming GeoTIFF data into 3D objects. Leveraging QGIS and Three.js streamlines the conversion process for integration into Unreal Engine. The resultant virtual reality application features distinct stations, enabling users to navigate, visualize, compare, and animate GIS data effectively. Each station caters to specific functionalities, ensuring a seamless and informative experience within the VR environment. This study also delves into augmented reality applications, adapting methodologies to address hardware limitations for smoother user experiences. By optimizing textures and implementing augmented reality functionalities through modules Swift, RealityKit, and ARKit, this study extends the immersive GIS experience to iOS devices. In conclusion, this research demonstrates the potential of integrating virtual reality, augmented reality, and GIS, pushing data visualization into new realms. The innovative workflows and applications developed serve as a testament to the evolving landscape of spatial data interpretation and engagement. [ABSTRACT FROM AUTHOR]

Learning to use geographic information system (GIS) software effectively may be intimidating due to the extensive range of features it offers. The GRASS GIS software, in particular, presents additional challenges for first-time users in terms of its complex startup procedure and unique terminology associated with its data structure. On the other hand, a substantial part of the GRASS user community including us as developers recognized and embraced the advantages of the current approach. Given the controversial nature of the whole issue, we decided to actively involve regular users by conducting several formal surveys and by performing usability testing. Throughout this process, we discovered that resolving specific software issues through pure user-centered design is not always feasible, particularly in the context of open-source scientific software where the boundary between users and developers is very fuzzy. To address this challenge, we adopted the user-centered methodology tailored to the requirements of open-source scientific software development, which we refer to as community-driven design. This paper describes the community-driven redesigning process on the GRASS GIS case study and sets a foundation for applying community-driven design in other open-source scientific projects by providing insights into effective software development practices driven by the needs and input of the project's community. [ABSTRACT FROM AUTHOR]

The accuracy of artificial intelligence techniques in estimating air quality is contingent upon a multitude of influencing factors. Unlike our previous study that examined PM2.5 over whole Europe using unbalanced spatial-temporal data, the focus of this study was on estimating PM2.5 specifically over the Czech Republic using more balanced dataset to train and evaluate the model. Moreover, the spatial autocorrelation between PM2.5 measurements was taken into consideration while building the model. The feature importance while developing the Extra Trees model revealed that spatial autocorrelation had greater significance in comparison to commonly used inputs such as elevation and NDVI. We found that R2 of the 10-CV for the new model was 16% higher than the previous one. Where R2 reached 0.85 with RMSE=5.42 µg/m3, MAE=3.41 µg/m3, and bias=-0.03 µg/m3. The developed spatiotemporal model was employed to generate comprehensive daily maps covering the entire study area throughout the period 2018-2020. The temporal analysis showed that the levels of PM2.5 exceeded recommended limits during the year 2018 in many regions. The eastern part of the country suffered from the highest concentrations especially over Zlín and Moravian-Silesian Regions. Air quality improved during the next two years in all regions reaching promising levels in 2020. The generated dataset will be available for other future air quality studies. [ABSTRACT FROM AUTHOR]

Vernacular architecture is an integral part of the national cultural heritage. Today, however, many of these buildings exist only on old plans or photographs and the average citizen has no opportunity to get acquainted with this part of the national identity. Therefore, in our work, we present the development of two web applications with the aim of creating a virtual open-air museum for presenting vernacular architecture in the Czech Republic. The applications were created using opensource technologies, and are implemented with methods that allow easy transfer from one operating system to another. The presented content is a carefully selected sub-sample of more than 10,000 available records representing all regional types of vernacular architecture. The result is one application designed for editors to manage the presented content and one application allowing interactive viewing of the available geo-located records designed for the general public. Individual records can be searched either directly using the map window or by querying the attribute table. These records contain descriptive information about the object, as well as historical photographs and plans and, for some objects, additional information in the form of 3D models, PDF documents and other files. The applications are designed in such a way that their content can be freely expanded in the future and thus contribute to the popularization of vernacular architecture among the general public, which was the main reason for their creation. [ABSTRACT FROM AUTHOR]

Air pollution is currently considered one of the most serious problems facing humans. Fine particulate matter with a diameter smaller than 2.5 micrometres (PM2.5) is a very harmful air pollutant that is linked with many diseases. In this study, we created a machine learning-based scheme to estimate PM2.5 using various open data such as satellite remote sensing, meteorological data, and land variables to increase the limited spatial coverage provided by ground-monitors. A space-time extremely randomised trees model was used to estimate PM2.5 concentrations over Europe, this model achieved good results with an out-of-sample cross-validated R2 of 0.69, RMSE of 5 μg/m3, and MAE of 3.3 μg/m3. The outcome of this study is a daily full coverage PM2.5 dataset with 1 km spatial resolution for the three-year period of 2018–2020. We found that air quality improved throughout the study period over all countries in Europe. In addition, we compared PM2.5 levels during the COVID-19 lockdown during the months March–June with the average of the previous 4 months and the following 4 months. We found that this lockdown had a positive effect on air quality in most parts of the study area except for the United Kingdom, Ireland, north of France, and south of Italy. This is the first study that depends only on open data and covers the whole of Europe with high spatial and temporal resolutions. The reconstructed dataset will be published under free and open license and can be used in future air quality studies. [ABSTRACT FROM AUTHOR]

The use of in situ references in Earth observation monitoring is a fundamental need. LUCAS (Land Use and Coverage Area frame Survey) is an activity that has performed repeated in situ surveys over Europe every three years since 2006. The dataset is unique in many aspects; however it is currently not available through a standardized interface, machine-to-machine. Moreover, the evolution of the surveys limits the performance of change analysis using the dataset. Our objective was to develop an open-source system to fill these gaps. This paper presents a developed system solution for the LUCAS in situ data harmonization and distribution. We have designed a multi-layer client-server system that may be integrated into end-to-end workflows. It provides data through an OGC (Open Geospatial Consortium) compliant interface. Moreover, a geospatial user may integrate the data through a Python API (Application Programming Interface) to ease the use in workflows with spatial, temporal, attribute, and thematic filters. Furthermore, we have implemented a QGIS plugin to retrieve the spatial and temporal subsets of the data interactively. In addition, the Python API includes methods for managing thematic information. The system provides enhanced functionality which is demonstrated in two use cases. [ABSTRACT FROM AUTHOR]

A spatiotemporal machine learning framework for automated prediction and analysis of long-term Land Use/Land Cover dynamics is presented. The framework includes: (1) harmonization and preprocessing of spatial and spatiotemporal input datasets (GLAD Landsat, NPP/VIIRS) including five million harmonized LUCAS and CORINE Land Cover-derived training samples, (2) model building based on spatial k-fold crossvalidation and hyper-parameter optimization, (3) prediction of the most probable class, class probabilities and model variance of predicted probabilities per pixel, (4) LULC change analysis on time-series of produced maps. The spatiotemporal ensemble model consists of a random forest, gradient boosted tree classifier, and an artificial neural network, with a logistic regressor as meta-learner. The results show that the most important variables for mapping LULC in Europe are: seasonal aggregates of Landsat green and near-infrared bands, multiple Landsat-derived spectral indices, long-term surface water probability, and elevation. Spatial cross-validation of the model indicates consistent performance across multiple years with overall accuracy (a weighted F1-score) of 0.49, 0.63, and 0.83 when predicting 43 (level-3), 14 (level-2), and five classes (level-1). Additional experiments show that spatiotemporal models generalize better to unknown years, outperforming single-year models on known-year classification by 2.7% and unknown-year classification by 3.5%. Results of the accuracy assessment using 48,365 independent test samples shows 87% match with the validation points. Results of time-series analysis (time-series of LULC probabilities and NDVI images) suggest forest loss in large parts of Sweden, the Alps, and Scotland. Positive and negative trends in NDVI in general match the land degradation and land restoration classes, with "urbanization" showing the most negative NDVI trend. An advantage of using spatiotemporal ML is that the fitted model can be used to predict LULC in years that were not included in its training dataset, allowing generalization to past and future periods, e.g. to predict LULC for years prior to 2000 and beyond 2020. The generated LULC time-series data stack (ODSE-LULC), including the training points, is publicly available via the ODSE Viewer. Functions used to prepare data and run modeling are available via the eumap library for Python. [ABSTRACT FROM AUTHOR]

Remote sensing is a useful tool for detecting landscape fragmentation, typically by creating land-use maps from remotely sensed images acquired at different dates. Nonetheless, classification may present a number of drawbacks since it degrades the information content of images leading to the loss of continuous information about fragmentation processes. For exploratory purposes, methods to detect landscape change based on continuous information should not require anya-prioriassumptions about landscape characteristics. Accordingly, Fourier transforms may represent the best algorithmic solution. In this paper, we describe a Fourier transform tool developed in a free and open-source environment to detect potential fragmentation over the landscape. We briefly introduce Fourier transforms applied to remotely sensed imagery by further showing their potential application with an empirical example. We argue that Fourier transforms represent a straightforward approach for detecting spatial fragmentation of the landscape, on the strength of their potential to detect trends in increase or decrease of complexity/heterogeneity of the landscape in an objective manner. To our knowledge, this is the first open-source tool for analysing fragmentation of the landscape in multitemporal series based on Fourier transforms, which guarantees a high robustness and reproducibility of the applied algorithms. [ABSTRACT FROM PUBLISHER]

Water erosion is the main cause of soil degradation in agricultural areas. Rill erosion can contribute vastly to the overall erosion rate. It is therefore crucial to identify areas prone to rill erosion in order to protect soil quality. Research on rainfall-runoff and subsequent sediment transport processes is often based on observing these processes at several scales, followed by a mathematical description of the observations. This paper presents the use of a combination of data obtained by different approaches at multiple scales to validate the SMODERP2D episodic hydrological-erosion model. This model describes infiltration, surface retention, surface runoff, and rill flow processes. In the model, the surface runoff generation is based on a water balance equation and is described by two separate processes: (a) for sheet flow, the model uses the kinematic wave approximation, which has been parameterized for individual soil textural classes using laboratory rainfall simulations, and (b) for rill flow, the Manning formula is used. Rill flow occurs if the critical water level of sheet flow is exceeded. The concept of model validation presented here uses datasets at different scales to study the surface runoff and erosion processes on the Býkovice agricultural catchment. The first dataset consisted of runoff generated by simulated rainfall on plots with dimensions of 2 × 8 m. The second dataset consisted of the runoff response to natural rainfall events obtained from long-term monitoring of 50 m2 plots. These two datasets were used to validate and calibrate the sheet flow and infiltration parameters. The third dataset consisted of occurrence maps of rills formed during heavy rainfalls obtained using remote sensing methods on a field plot with an area of 36.6 ha. This last dataset was used to validate the threshold critical water level that is responsible in the model for rill flow initiation in the SMODERP2D model. The validation and the calibration of the surface runoff are performed well according to the Nash–Sutcliffe efficiency coefficient. The scale effect was evident in the 50 m2 plots where parameters lower than the mean best fit the measured data. At the field plot scale, pixels with measured rills covered 5% of the total area. The best model solution achieved a similar rill cover for a vegetated soil surface. The model tended to overestimate the occurrence of rills in the case of simulations with bare soil. Although rills occurred both in the model and in the monitored data in many model runs, a spatial mismatch was often observed. This mismatch was caused by flow routing algorithm displacement of the runoff path. The suitability of the validation and calibration process at various spatial scales has been demonstrated. In a future study, data will be obtained from various localities with various land uses and meteorological conditions to confirm the transferability of the procedure. [ABSTRACT FROM AUTHOR]

The recent COVID-19 pandemic affected various aspects of life. Several studies established the consequences of pandemic lockdown on air quality using satellite remote sensing. However, such studies have limitations, including low spatial resolution or incomplete spatial coverage. Therefore, in this paper, we propose a machine learning-based scheme to solve the pre-mentioned limitations by training an optimized space-time extra trees model for each year of the study period. The results have shown that our trained models reach a prediction accuracy up to 95% when predicting the missing values in the MODIS MCD19A2 Aerosol Optical Depth (AOD) product. The outcome of the mentioned scheme was a geo-harmonized atmospheric dataset for aerosol optical depth at 550 nm with 1 km spatial resolution and full coverage over Europe. As an application, we used the proposed machine learning based prediction approach in AOD levels analysis. We compared the mean AOD levels between the lockdown period from March to June in 2020 and the mean AOD values of the same period for the past 5 years. We found that AOD levels dropped over most European countries in 2020 but increased in several eastern and western countries. The Netherlands had the most significant average decrease in AOD levels (19%), while Spain had the highest average increase (10%). Moreover, we analyzed the relationship between the relative percentage difference of AOD and four meteorological variables. We found a positive correlation between AOD and relative humidity and a negative correlation between AOD and wind speed. The value of the proposed prediction scheme is further emphasized by taking into consideration that the reconstructed dataset can be used for future air quality studies concerning Europe. [ABSTRACT FROM AUTHOR]

Hydrological modelling of the runoff response to the causal rainfall on the small catchment or individual plot scale is an important part of designing the water management measures in the landscape. Classic methods based on CN and unit hydrograph are implemented in number of models (HEC-HMS, WMS, etc.). This approach is also being implemented in the Atlas environment. Expanding the use of physically-based approach still faces number of ambiguities. Obtaining parameters of hydrological models may be a difficult task. Measurement of physically based parameters is often time consuming and costly. On the other side, obtaining parameters by calibration require long term time series of data which are often not available in appropriate quality. In contrast to other soil properties, soil texture is relatively easy to obtain. In this contribution we present an attempt to link soil texture to surface runoff model parameters. Serie of artificial rainfall experiments on 4 m2 plot with disturbed soil of various texture was undertaken. Besides soil texture the experiments differ in terms of rainfall intensity and slope. Episodic rainfall-runoff model smoderp2d was used to model those experiments. Parameters of Philip's infiltration, required by the model, were calculated directly for each single experiment. Manning-Strickler formula is used in simplified solution of the kinematic wave equation to express the momentum. Manning-Strickler (MS) formula, which is usually used to 1D channel flow calculations, is used to model the surface sheet flow. Roughness coefficient, which express the effect of current state of soil aggregates, texture and vegetation, is important parameter of MS formula. Roughness coefficient and other two parameters, related to surface slope and water level height were optimised. Response sensitivity of the runoff on these parameters is presented. The set of parameters were further used in infer uncertainty in runoff for a given soil texture class. In addition, we would like to present the new development of the model smoderp2d which were recently integrated in the GRASS GIS and QGIS open source software packages.The research has been supported by the research grants SGS17/173/OHK1, TJ01000270 and QK1910029. [ABSTRACT FROM AUTHOR]