Large Geospatial Datasets Research Articles

AI-Enhanced Geomatics Engineering: Innovative Solutions and Applications Using ChatGPT, an Advanced AI Language Model

This study investigates the potential of the advanced artificial intelligence language model, ChatGPT, in innovative solutions and applications within Geomatics Engineering, under the title "AI-Enhanced Geomatics Engineering: Innovative Solutions and Applications Using ChatGPT". ChatGPT is renowned for its ability to generate human-like text and has been trained on extensive datasets. Geomatics Engineering encompasses various sub-disciplines such as geographic information systems (GIS), remote sensing, global positioning systems (GPS), geodesy, and geomatics engineering. The research analyzes how AI technology, exemplified by models like ChatGPT, can be leveraged in Geomatics Engineering projects to enhance project management, improve data analysis accuracy, increase process efficiency, and impact project outcomes. AI models like ChatGPT enable the analysis of large geospatial datasets and play a crucial role in understanding and interpreting complex datasets. Furthermore, these technologies are expected to save time and contribute to obtaining more accurate results in Geomatics Engineering applications. The study also demonstrates how the integration of AI into Geomatics Engineering disciplines can be achieved through case studies and practical applications. These case studies illustrate how AI-supported technologies can be applied to real-world problems and improve current practical applications within the industry. In conclusion, this research underscores how AI can play a progressive role in Geomatics Engineering and emphasizes the transformative impact future technological advancements can have in these disciplines. The integration and utilization of such technologies are of paramount importance for exploring potential applications in this field.

Machine learning and deep learning-based landslide susceptibility mapping using geospatial techniques in Wayanad, Kerala state, India

Landslide susceptibility mapping is vital for disaster management and sustainable land-use planning. This research was conducted in Wayanad, Kerala, India, to identify landslide susceptible zones. The study used large geospatial datasets, such as elevation, slope, aspect, curvature, stream power index, topographic wetness index, land use and land cover, rainfall, flow accumulation, geology, and geomorphology. It is followed by the application of various machine learning and deep learning models such as the support vector machine, artificial neural networks, logistic regression, random forest, gradient boosting machine, recurrent neural networks long short-term memory, and deep neural network models to map the landslide susceptible zones. The model was trained and validated using the landslide inventory map, which contains 298 sites of landslides. The random forest model, with 97 % accuracy, performed best. It is possible to effectively mitigate landslides and plan long-term land use by identifying hazardous zones within the study region.

Estimating global trends of air pollution, air pollution-attributable disease burdens, and CO2 emissions in 13,000 cities using large geospatial datasets

Estimating global trends of air pollution, air pollution-attributable disease burdens, and CO2 emissions in 13,000 cities using large geospatial datasets

Remote sensing for rangeland conservation monitoring: Impacts of livestock removal after 15 years

Abstract Degradation from overgrazing of global arid rangelands is of significant concern for this widespread terrestrial ecosystem, and the people who rely on its sustainable use. Addressing this degradation requires a better understanding of how livestock removal and conservation management actions affect the rehabilitation of land condition. However, the dominant effect of climate and meteorological variability on arid vegetation makes isolating management‐induced change a difficult task. We present an analytically simple method for measuring relative land condition by assessing differences in remote‐sensing‐derived persistent fractional vegetation between adjacent stocked and destocked regions over time. The 22‐year observation period focusses on vegetation recovery at a long‐established sheep station after the removal of livestock and the introduction of conservation‐focussed management. The study region, in the southern Australian arid rangelands, comprises sparsely vegetated chenopod shrublands and acacia woodlands across low‐lying plains. Comparison is made to immediately surrounding stocked properties in a spatial design where climate and biogeographic variables remain consistent across the study area. We found an unequivocal relative increase in vegetation cover, and reduction of bare ground cover at the destocked property that became evident approximately 8 years after a reduction in grazing pressure. Distinctly higher persistent vegetation cover was maintained, even during periods of drought. Non‐photosynthetic vegetation cover was most indicative of long‐term change and increased by 1.1% relative to surrounding areas. Photosynthetic vegetation cover increased by 0.5%, while bare ground cover decreased by 2.1%. These cover changes, particularly in the non‐photosynthetic vegetation and bare ground components spatially coincide with the fenced boundary of the destocked property. Synthesis and applications: The ability to quantify management‐related impacts in land condition using freely available remote sensing data could support improved management by all stakeholders in these regions. The method leverages readily available satellite remote sensing data in a study design that is not dependent on large geospatial and climatic datasets that are not necessarily available in remote regions.

Executing Ad-Hoc Queries on Large Geospatial Data Sets Without Acceleration Structures

In this case study, we investigate if it is possible to harness the capabilities of modern commodity hardware to perform ad-hoc queries on large raw geospatial data sets. Normally, this requires building an index structure, which is a time-consuming process. We aim to provide means to individual users who receive a new or updated geospatial data set and want to directly start working with it without having to build such an index structure first. To this end, we conduct various experiments on two distinct types of data: 3D building models and point clouds. For the former, we demonstrate that well-known algorithms such as fast string search allow a wide range of queries to be answered in at most a few seconds on data sets with over a million buildings. The usage of progressive indexing additionally improves query run time by more than a factor of two. Regarding point clouds, we achieve similar run times using the popular LAS file format and a query throughput of up to a billion points per second when using a columnar memory layout. The run time of ad-hoc queries is often on par with that of database-driven solutions, sometimes even outperforming them. Considering that ad-hoc queries require no preprocessing, our results show that they are a viable alternative to acceleration structures when working with geospatial data.

HTF: Homogeneous Tree Framework for Differentially-Private Release of Large Geospatial Datasets with Self-Tuning Structure Height.

Mobile apps that use location data are pervasive, spanning domains such as transportation, urban planning and healthcare. Important use cases for location data rely on statistical queries, e.g., identifying hotspots where users work and travel. Such queries can be answered efficiently by building histograms. However, precise histograms can expose sensitive details about individual users. Differential privacy (DP) is a mature and widely-adopted protection model, but most approaches for DP-compliant histograms work in a data-independent fashion, leading to poor accuracy. The few proposed data-dependent techniques attempt to adjust histogram partitions based on dataset characteristics, but they do not perform well due to the addition of noise required to achieve DP. In addition, they use ad-hoc criteria to decide the depth of the partitioning. We identify density homogeneity as a main factor driving the accuracy of DP-compliant histograms, and we build a data structure that splits the space such that data density is homogeneous within each resulting partition. We propose a self-tuning approach to decide the depth of the partitioning structure that optimizes the use of privacy budget. Furthermore, we provide an optimization that scales the proposed split approach to large datasets while maintaining accuracy. We show through extensive experiments on large-scale real-world data that the proposed approach achieves superior accuracy compared to existing approaches.

Geospatial Analysis of Alaskan Lakes Indicates Wetland Fraction and Surface Water Area Are Useful Predictors of Methane Ebullition

Arctic-boreal lakes emit methane (CH4), a powerful greenhouse gas. Recent studies suggest ebullition might be a dominant methane emission pathway in lakes but its drivers are poorly understood. Various predictors of lake methane ebullition have been proposed but are challenging to evaluate owing to different geographical characteristics, field locations, and sample densities. Here we compare large geospatial data sets of lake area, lake perimeter, permafrost, land cover, temperature, soil organic carbon content, depth, and greenness with remotely sensed methane ebullition estimates for 5,143 Alaskan lakes. We find that lake wetland fraction (LWF), a measure of lake wetland and littoral zone area, is a leading predictor of methane ebullition (adj. R 2 = 0.211), followed by lake surface area (adj. R 2 = 0.201). LWF is inversely correlated with lake area, thus higher wetland fraction in smaller lakes might explain a commonly cited inverse relationship between lake area and methane ebullition. Lake perimeter (adj. R 2 = 0.176) and temperature (adj. R 2 = 0.157) are moderate predictors of lake ebullition, and soil organic carbon content, permafrost, lake depth, and greenness are weak predictors. The low adjusted R 2 values are typical and informative for methane attribution studies. Our leading model, which uses lake area, temperature, and LWF (adj. R 2 = 0.325, n = 5,130) performs slightly better than leading multivariate models from similar studies. Our results suggest landscape-scale geospatial analyses can complement smaller field studies, for attributing Arctic-boreal lake methane emissions to readily available environmental variables.

Quantitative Morphometric 3D Terrain Analysis of Japan Using Scripts of GMT and R

In this paper, we describe two related scripting methods of cartographic data processing and visualization that provide 2D and 3D mapping of Japan with different algorithm complexity. The first algorithm utilizes Generic Mapping Toolset (GMT), which is known as an advanced console-based program for spatial data processing. The modules of GMT combine the functionality of scripting with the aspects of geoinformatics, which is especially effective for the rapid analysis of large geospatial datasets, multi-format data processing, and mapping in 2D and 3D modes. The second algorithm presents the use of the R programming language for cartographic visualization and spatial analysis. This R method utilizes the packages ‘tmap’, ‘raster’, ‘maps’, and ‘mapdata’ to model the morphometric elements of the Japanese archipelago, such as slope, aspect, hillshade and elevation. The general purpose graphical package ‘ggplot2’ of R was used for mapping the prefectures of Japan. The two scripting approaches demonstrated an established correspondence between the programming languages and cartography determined with the use of scripts for data processing. They outperform several well-known and state-of-the-art GIS methods for mapping due to their high automation of data processing. Cartography has largely reflected recent advances in data science, the rapid development of scripting languages, and transfer in the approaches of data processing. This extends to the shift from the traditional GIS to programming languages. As a response to these new challenges, we demonstrated in this paper the advantages of using scripts in mapping, which consist of repeatability and the flexible applicability of scripts in similar works.

Interactive analysis of big geospatial data with high‐performance computing: A case study of partisan segregation in the United States

AbstractResearchers are increasingly working with large geospatial datasets that contain hundreds of millions of records. At this scale, desktop GIS systems typically fall short and so new approaches and methods are needed. The objective of this work is to develop new approaches to interactively analyze large datasets and then to demonstrate the usefulness of those approaches using a case study looking at voter, or partisan segregation. Historically, the measurement of partisan segregation has been limited to comparing large geographic areas such as counties or states because researchers only had access to aggregated data. In this case study, however, we measure partisan segregation down to the individual for 180 million U.S. voters using advanced geospatial data science and high‐performance computing. This article discusses interactive method development for big geospatial data analysis including techniques used, solutions developed, and processing time statistics.

Review on Integrating Geospatial Big Datasets

Abstract: Big Data and Local Systems (GIS) are both expanding technologies has influenced many areas over the past 10 years and will continue to improve and help resolve a global crisis problems, such as the effects of climate change or a global epidemic. More GIS applications it works with the continued growth of large geospatial data sources to drive precise and informed decisions. Geospatial Big Data Integration is designed to achieve compatibility with different geospatial data sets without the availability of space. A large number of large geospatial data sources seek to operate effectively the integration of data storage and management of such data, which will be used for geospatial data analysis and vision. For example, risk management data sets are related to health care and the environment heterogeneous and distinct. Finding an integrated view of large geospatial databases is also difficult it is challenging, especially when we consider the problems associated with the health epidemic and natural disasters. Therefore, before we try to predict and mitigate the processes that take place in these domains, we should see that big data integration is very important in combining data sets. We explore and chat issues involved in compiling large geospatial data sets in this study. We then split the big data collection it processes it in three phases, namely, data storage, data conversion and integration methods. In addition, several research challenges focus on large geospatial data, large global data, data retention, data conversion and linked data are presented. Lastly, open up research issues and emerging styles requires in-depth research into the near future highlighted in this study. Keywords: Big Data, Geospatial

Nearest‐neighbor sparse Cholesky matrices in spatial statistics

AbstractGaussian process (GP) is a staple in the toolkit of a spatial statistician. Well‐documented computing roadblocks in the analysis of large geospatial datasets using GPs have now largely been mitigated via several recent statistical innovations. Nearest neighbor Gaussian process (NNGP) has emerged as one of the leading candidates for such massive‐scale geospatial analysis owing to their empirical success. This article reviews the connection of NNGP to sparse Cholesky factors of the spatial precision (inverse‐covariance) matrix. Focus of the review is on these sparse Cholesky matrices which are versatile and have recently found many diverse applications beyond the primary usage of NNGP for fast parameter estimation and prediction in the spatial (generalized) linear models. In particular, we discuss applications of sparse NNGP Cholesky matrices to address multifaceted computational issues in spatial bootstrapping, simulation of large‐scale realizations of Gaussian random fields, and extensions to nonparametric mean function estimation of a GP using random forests. We also review a sparse‐Cholesky‐based model for areal (geographically aggregated) data that addresses long‐established interpretability issues of existing areal models. Finally, we highlight some yet‐to‐be‐addressed issues of such sparse Cholesky approximations that warrant further research.This article is categorized under:Algorithms and Computational Methods > AlgorithmsAlgorithms and Computational Methods > Numerical Methods

Forest Fire Mapping using Normalized Burned Ratio and Cloud Computing to Calculate the Losses Incurred in Mount Lawu, Magetan Regency

Forest and land fires in Indonesia are one of the most common disasters. Over time, Indonesia’s forests have experienced a decline, one of which is the result of forest and land fires. Forest and land fires can occur due to temperatures in the dry season that continue to increase. Utilization of remote sensing technology can be used for mapping forest and fire areas, and can also be used as a reference for the government to carry out reforestation or reforestation processes in fire areas. The use of Multispectral Sensors in Sentinel 2 satellite imagery can be used to produce this research. This research takes a case study on Mount Lawu, Magetan Regency. Mount Lawu is one of the climbing tourist destinations that almost every year experiences forest and land fires. And the use of the Google Earth Engine platform that supports cloud computing to facilitate this research in accessing high data sources to process large geospatial data sets. For processing satellite images using the Normalized Burned Ratio (NBR) vegetation index processing algorithm. This research was conducted in 2018 and 2019. By using the Random Forest method, which is one of the applications of Machine Learning, it will produce several accurate results from the calculation results. The results of the calculation of the area of fire in 2018 were 284.88 hectares with a total loss of Rp 42,732,000.00 and in 2019 it was 146.03 hectares with a total loss of Rp 21,904,500. The results of this study are displayed on GEE Apps so that users can directly access the resulting data.

Using Machine Learning to Map Western Australian Landscapes for Mineral Exploration

Landscapes evolve due to climatic conditions, tectonic activity, geological features, biological activity, and sedimentary dynamics. Geological processes at depth ultimately control and are linked to the resulting surface features. Large regions in Australia, West Africa, India, and China are blanketed by cover (intensely weathered surface material and/or later sediment deposition, both up to hundreds of metres thick). Mineral exploration through cover poses a significant technological challenge worldwide. Classifying and understanding landscape types and their variability is of key importance for mineral exploration in covered regions. Landscape variability expresses how near-surface geochemistry is linked to underlying lithologies. Therefore, landscape variability mapping should inform surface geochemical sampling strategies for mineral exploration. Advances in satellite imaging and computing power have enabled the creation of large geospatial data sets, the sheer size of which necessitates automated processing. In this study, we describe a methodology to enable the automated mapping of landscape pattern domains using machine learning (ML) algorithms. From a freely available digital elevation model, derived data, and sample landclass boundaries provided by domain experts, our algorithm produces a dense map of the model region in Western Australia. Both random forest and support vector machine classification achieve approximately 98% classification accuracy with a reasonable runtime of 48 minutes on a single Intel® Core™ i7-8550U CPU core. We discuss computational resources and study the effect of grid resolution. Larger tiles result in a more contiguous map, whereas smaller tiles result in a more detailed and, at some point, noisy map. Diversity and distribution of landscapes mapped in this study support previous results. In addition, our results are consistent with the geological trends and main basement features in the region. Mapping landscape variability at a large scale can be used globally as a fundamental tool for guiding more efficient mineral exploration programs in regions under cover.

Visualization of Big Spatial Data using Coresets for Kernel Density Estimates.

The size of large, geo-located datasets has reached scales where visualization of all data points is inefficient. Random sampling is a method to reduce the size of a dataset, yet it can introduce unwanted errors. We describe a method for subsampling of spatial data suitable for creating kernel density estimates from very large data and demonstrate that it results in less error than random sampling. We also introduce a method to ensure that thresholding of low values based on sampled data does not omit any regions above the desired threshold when working with sampled data. We demonstrate the effectiveness of our approach using both, artificial and real-world large geospatial datasets.

Regression Tree Based Correlation Technique in Spatial Data Classification

Data mining is the process of discovering useful patterns from large geo-spatial datasets with the help of machine learning methods. . The machine learning methods plays an important role for data analytics modeling and visualization. Geo-spatial data is a significant task in many application domains, such as environmental science, geographic information science, and social networks. However, the existing spatial pattern discovery and prediction techniques failed to predict the event accurately with minimum error and time consumption. . In this paper, a novel Pearson Correlated Regression Tree-based Affine Projective spatial data Classification (PCRT-APSDC) technique is proposed to improve the spatial data classification and minimize error based on the Affine Projective classification technique. The proposed algorithm employs a fuzzy rule procedure that constructs the regression tree. The fuzzy rule is applied for linking the inputs (i.e. spatial data) with the outputs (i.e. classification results). Our goal is to classify the data into two subsets such as fired region and non-fired region. Experimental evaluation is carried out using a forest fire dataset with different factors such as classification accuracy, false-positive rate, and classification time. The results confirm that the proposed technique predicts the fired region with increased spatial data classification accuracy and minimized time as well as false-positive rate than the state-of-the-art methods.

Can-GLWS: Canadian Great Lakes Weather Service for the Soil and Water Assessment Tool (SWAT) modelling

The rapid rise in availability of large geospatial datasets for the development of hydrological models such as Soil and Water Assessment Tool (SWAT) has led to a dramatic increase in both the demand and availability of web services and tools that assist watershed modellers in incorporating data and knowledge into their modelling frameworks. Within the Canadian Great Lakes region, there is a huge potential for the application of SWAT in integrated water resources management. However, a potential barrier is the preparation of SWAT weather inputs that require time-intensive preprocessing of large data sets. Because such preprocessing is reproducible, the redundancy associated with it can be removed by introducing a web service that enables easy and open dissemination of climate data (including climate change and historical data) in SWAT-ready format. This short communication introduces such a web service called the Canadian Great Lakes Weather Data Service for SWAT (Can-GLWS). It hosts observed (historical) and projected (future) daily precipitation, daily maximum/minimum temperature, as well as weather generator database at regular grids (300 arc seconds or ~10 km) for use in SWAT simulations of the region. The novel Can-GLWS web service offers flexibility in selecting the region of interest by allowing them to be uploaded as a shapefile or to draw a rectangle or a polygon. We believe that such data as a service platform will help many practitioners to explore several issues pertaining to the sustainability of the freshwater resources of Canadian Great Lakes using the SWAT model.

Interactive data styling and multifocal visualization for a multigrid web-based Digital Earth

ABSTRACT Globe-based Digital Earth (DE) is a promising system that uses 3D models of the Earth for integration, organization, processing, and visualization of vast multiscale geospatial datasets. The growing size and scale of geospatial datasets present significant obstacles to interactive viewing and meaningful visualizations of these DE systems. To address these challenges, we present a novel web-based multiresolution DE system using a hierarchical discretization of the globe on both server and client sides. The presented web-based system makes use of a novel data encoding technique for rendering large multiscale geospatial datasets, with the additional capability of displaying multiple simultaneous viewpoints. Only the data needed for the current views and scales are encoded and processed. We leverage the power of GPU acceleration on the client-side to perform real-time data rendering and dynamic styling. Efficient rendering of multiple views allows us to support multilevel focus+context visualization, an effective approach to navigate through large multiscale global datasets. The client–server interaction as well as the data encoding, rendering, styling, and visualization techniques utilized by our presented system contribute toward making DE more accessible and informative.

An invader in salmonid rearing habitat: current and future distributions of smallmouth bass (Micropterus dolomieu) in the Columbia River Basin

Invasive species and climate change are leading threats to freshwater ecosystems. In the Columbia River Basin (CRB), nonnative fishes are a critical consideration in salmon recovery, yet managers lament a lack of distribution information. Combining a species distribution model (SDM) with environmental DNA (eDNA), we locate range boundary regions of nonnative smallmouth bass (Micropterus dolomieu) and evaluate its overlap with native salmonids. A combination of thermal, hydrological, and geomorphic variables predict that smallmouth bass is distributed across ∼18 000 river kilometres and overlaps with 3%–62% of rearing habitat of salmonids (species-dependent) in the CRB. Under a moderate climate change scenario, smallmouth bass is predicted to expand its range by two-thirds (totaling ∼30 000 river kilometres) by 2080. Basin-wide models were sufficiently accurate to identify upstream invasion extents to within 15 km of the eDNA-based boundary, and including eDNA data improved model performance at critical range boundary regions without sacrificing broadscale model performance. Our study highlights how eDNA approaches can supplement large geospatial data sets to result in more accurate SDM predictions, guiding nonnative species management.

Understanding Collective Human Mobility Spatiotemporal Patterns on Weekdays from Taxi Origin-Destination Point Data.

With the availability of large geospatial datasets, the study of collective human mobility spatiotemporal patterns provides a new way to explore urban spatial environments from the perspective of residents. In this paper, we constructed a classification model for mobility patterns that is suitable for taxi OD (Origin-Destination) point data, and it is comprised of three parts. First, a new aggregate unit, which uses a road intersection as the constraint condition, is designed for the analysis of the taxi OD point data. Second, the time series similarity measurement is improved by adding a normalization procedure and time windows to address the particular characteristics of the taxi time series data. Finally, the DBSCAN algorithm is used to classify the time series into different mobility patterns based on a proximity index that is calculated using the improved similarity measurement. In addition, we used the random forest algorithm to establish a correlation model between the mobility patterns and the regional functional characteristics. Based on the taxi OD point data from Nanjing, we delimited seven mobility patterns and illustrated that the regional functions have obvious driving effects on these mobility patterns. These findings are applicable to urban planning, traffic management and planning, and land use analyses in the future.

Mapping up-to-Date Paddy Rice Extent at 10 M Resolution in China through the Integration of Optical and Synthetic Aperture Radar Images

Rice is a staple food in East Asia and Southeast Asia—an area that accounts for more than half of the world’s population, and 11% of its cultivated land. Studies on rice monitoring can provide direct or indirect information on food security, and water source management. Remote sensing has proven to be the most effective method for the large-scale monitoring of croplands, by using temporary and spectral information. The Google Earth Engine (GEE) is a cloud-based platform providing access to high-performance computing resources for processing extremely large geospatial datasets. In this study, by leveraging the computational power of GEE and a large pool of satellite and other geophysical data (e.g., forest and water extent maps, with high accuracy at 30 m), we generated the first up-to-date rice extent map with crop intensity, at 10 m resolution in the three provinces with the highest rice production in China (the Heilongjiang, Hunan and Guangxi provinces). Optical and synthetic aperture radar (SAR) data were monthly and metric composited to ensure a sufficient amount of up-to-date data without cloud interference. To remove the common confounding noise in the pixel-based classification results at medium to high resolution, we integrated the pixel-based classification (using a random forest classifier) result with the object-based segmentation (using a simple linear iterative clustering (SLIC) method). This integration resulted in the rice planted area data that most closely resembled official statistics. The overall accuracy was approximately 90%, which was validated by ground crop field points. The F scores reached 87.78% in the Heilongjiang Province for monocropped rice, 89.97% and 80.00% in the Hunan Province for mono- and double-cropped rice, respectively, and 88.24% in the Guangxi Province for double-cropped rice.