Spatial Entities Research Articles

Achieving fine-grained urban flood perception and spatio-temporal evolution analysis based on social media

Timely understanding of affected areas during disasters is essential for the implementation of emergency response activities. As one of the low-cost and information-rich volunteer geographic information, social media data can reflect geographic events through human behavior, which is a powerful supplementary source for fine-grained flood monitoring in urban areas. However, the value of social media data has not been fully exploited as potential location and water depth information may be embedded in both text and images. In this study, we propose a novel framework for fine-grained information extraction and dynamic spatial-temporal awareness in disaster-stricken areas based on Sina Weibo. First, we construct a novel fine-grained location corpus specifically for urban flooding contexts. The corpus summarizes characteristics of address descriptions in flood-related Weibo texts, including standard address entities and spatial relationship entities, based on the named entity recognition (NER) model. Then, water depth information in texts and images is obtained based on different deep learning modules and fused at the decision level. Specifically, in text analysis module, we summarize and extract diverse descriptions of water depth, and in image analysis module, we develop a water level hierarchical mapping method. Finally, we analyze the spatio-temporal distribution characteristics and variation patterns of the extracted information to enhance situational awareness. Taking the urban flood occurred in Anhui, China as a case study, we find that the variation of flooding hotspot areas in Sina Weibo and rainfall centers show a significant spatial and temporal consistency, and the fusion of text and image-based information can facilitate dynamic perception of flood processes. The framework presented in this study provides a feasible way to implement refined situational awareness and spatio-temporal evolution analysis of urban floods at the city level in time.

Global optimisation matching method for multi-representation buildings constrained by road network

Entity matching is one of the key technologies for geospatial data update and fusion. In response to the shortcomings of most spatial entity matching methods that use local optimisation strategies, a global optimisation matching method for multi-representation buildings using road network constraints is proposed. First, the road network is used for region segmentation to obtain candidate matches. Second, the spatial similarity among the candidate matching objects is calculated and the characteristic similarity weights are determined using the entropy weight method. Third, the matching of building entities is transformed into an allocation problem using integer programming ideas, and the Hungarian algorithm is solved to obtain the optimal matching combination with minimum global variance. Finally, two test areas are selected to validate the proposed method, and the precision, recall, and F-measure values of the experiments are 96.35%, 97.11%, and 96.73% versus 95.96%, 97.03%, and 96.49%, respectively. The matching accuracy is greatly improved compared with the local search strategy.

Knowledge Spillover and Spatial Innovation Growth: Evidence from China’s Yangtze River Delta

This article explores the relationship between knowledge sources at different levels and corporate innovation from the perspective of urban cluster, with a focus on enterprises. This paper conducted an empirical analysis of 375 listed companies in 27 cities within the Yangtze River Delta urban cluster in China from 2009 to 2019. The findings showed that: (1) Local scientific knowledge spillovers, mediated by industry relevance, positively influence firms’ innovation performance. This study verifies how spatial knowledge is dimensionally reduced from scientific spillovers to industrial technological innovation. (2) Emerging industries acquire relevant scientific knowledge for transformation from a broader regional scope. Regional knowledge creation in the Yangtze River Delta urban cluster has stimulated industrial innovation across various sectors, thereby enhancing the overall innovation capacity and level of the urban cluster. (3) Regional diversity significantly affects the process of transforming knowledge into innovation. This paper supports the existence of a unified spatial innovation network among heterogeneous spatial economic entities and emphasizes the innovation synergy from lower to higher levels within heterogeneous hierarchical innovation networks. Developing urban agglomeration strategies that leverage the resource advantages of industrial clusters and adjust industrial layouts is an important approach to promote innovation and economic growth.

Spatio-Temporal Information Extraction and Geoparsing for Public Chinese Resumes

As an important carrier of individual information, the resume is an important data source for studying the spatio-temporal evolutionary characteristics of individual and group behaviors. This study focuses on spatio-temporal information extraction and geoparsing from resumes to provide basic technical support for spatio-temporal research based on resume text. Most current studies on resume text information extraction are oriented toward recruitment work, such as the automated information extraction, classification, and recommendation of resumes. These studies ignore the spatio-temporal information of individual and group behaviors implied in resumes. Therefore, this study takes the public resumes of teachers in key universities in China as the research data, proposes a set of spatio-temporal information extraction solutions for electronic resumes of public figures, and designs a spatial entity geoparsing method, which can effectively extract and spatially locate spatio-temporal information in the resumes. To verify the effectiveness of the proposed method, text information extraction models such as BiLSTM-CRF, BERT-CRF, and BERT-BiLSTM-CRF are selected to conduct comparative experiments, and the spatial entity geoparsing method is verified. The experimental results show that the precision of the selected models on the named entity recognition task is 96.23% and the precision of the designed spatial entity geoparsing method is 97.91%.

An Ontology-Based Framework for Geospatial Integration and Querying of Raster Data Cube Using Virtual Knowledge Graphs

The integration of the raster data cube alongside another form of geospatial data (e.g., vector data) raises considerable challenges when it comes to managing and representing it using knowledge graphs. Such integration can play an invaluable role in handling the heterogeneity of geospatial data and linking the raster data cube to semantic technology standards. Many recent approaches have been attempted to address this issue, but they often lack robust formal elaboration or solely concentrate on integrating raster data cubes without considering the inclusion of semantic spatial entities along with their spatial relationships. This may constitute a major shortcoming when it comes to performing advanced geospatial queries and semantically enriching geospatial models. In this paper, we propose a framework that can enable such semantic integration and advanced querying of raster data cubes based on the virtual knowledge graph (VKG) paradigm. This framework defines a semantic representation model for raster data cubes that extends the GeoSPARQL ontology. With such a model, we can combine the semantics of raster data cubes with features-based models that involve geometries as well as spatial and topological relationships. This could allow us to formulate spatiotemporal queries using SPARQL in a natural way by using ontological concepts at an appropriate level of abstraction. We propose an implementation of the proposed framework based on a VKG system architecture. In addition, we perform an experimental evaluation to compare our framework with other existing systems in terms of performance and scalability. Finally, we show the potential and the limitations of our implementation and we discuss several possible future works.

Examining the representation of landscape and its emotional value in German-Swiss fiction between 1840 and 1940

This paper presents a quantitative analysis of the representation and affective encoding of fictional space in a corpus of 125 Swiss literary prose texts of the 19th and early 20th Century written in German, offering a contribution to both spatial and affective literary studies. Motivated by questions about the iconic dichotomy between ‘urban’ and ‘rural/natural’ space in literary works (Sengle; Fournier; Nell and Weiland) – and in Swiss literature around 1900 in particular (Rehm) – we use computational methods to detect and examine how different types of space are distributed and affectively encoded in German-Swiss literature. Taking into account the complexity of cultural perceptions and representations of space across history, we examine the presence of ‘urban’ and ‘rural/natural’ fictional spaces and their potential role in constructing a ‘Swiss’ national literature (Böhler; Zimmer), and their affective encoding. In order to do this, we first compiled a comprehensive dictionary of named and non-named spatial entities in the broad spatial categories RURAL and URBAN, and examined the presence of sentiment and emotions (valence and discrete emotions) and their ‘strength’ (arousal) in relation to these. We used current state-of-the-art sentiment lexicons for German available to the digital humanities community. Similarly to Heuser et al., we mapped the spatial entities and the sentiment lexicons onto our corpus, and focused on spans of +/-50 words around the detected entities, in order to examine the specific sentiment and emotions related to space. In an exploratory analysis, we offer here a first-time data-driven perspective on rural and urban fictional space, incorporating the dimension of affective encoding of space systematically.

Geospatial Tool Development for the Management of Historical Hiking Trails—The Case of the Holy Site of Meteora

This paper presents a holistic guiding methodology for the development of a geospatial tool to be used for the documentation, planning, smart management and dissemination of a country’s network of historic hiking trails. To deal with the challenges and to ensure the sustainability of a historic site, geospatial documentation merging authoritative and crowdsourced data and a WebGIS-based spatial analysis is necessary. Geospatial data collection should include professional field surveys, professional and crowdsourced photographic documentation and video recording of the existing historic walking/hiking trails. A geodatabase, structured using relational model technology, including vector spatial entities (feature classes), mosaics (raster) and tabulated data (geodatabase tables), should be developed on a commercial or open platform; in this case, the ArcGIS Pro is used. Entities with embedded descriptive information and metadata for the technical, legal, historical, and administrative context may then be created. An object-oriented data model is needed to connect spatial and descriptive information. Spatial and descriptive queries or correlations between attribute fields of spatial entities must be enabled for specialized information retrieval by either experts or users. Next, a web GIS application to present the developed geodatabase in a visually appealing and informative way is created. It should integrate 2D maps with built-in tools and should support advanced functionalities, such as: (i) pop-ups that display brief information and images about specific spots along the trails; (ii) dynamic visualization of the vertical profile of each trail; (iii) multimedia information about landmarks, natural features and scenic viewpoints. Finally, the tool includes a feedback service and continuous efficiency monitoring and assessment, and enables adjustments, if and where needed. The tool is tested and used for 10 historical walking/hiking trails of the archaeological and Holy Site of Meteora, Central Greece. This is a UNESCO World Heritage site. The network, with a total length of 35 km, leads to six monasteries, still active since the 12th century, passing by gigantic rocks and beautiful natural landscapes. The site is famous globally and the greater area is continuously overcrowded with visitors. The tool is anticipated to be used for the documentation and management of the whole walking/hiking historic trail network of Greece in the future.

Mining Geospatial Relationships from Text

A geospatial Knowledge Graph (KG) is a heterogeneous information network, capable of representing relationships between spatial entities in a machine-interpretable format, and has tremendous applications in logistics and social networks. Existing efforts to build a geospatial KG, have mainly used sparse spatial relationships, e.g., a district located inside a city, which provide only marginal benefits compared to a traditional database. In spite of the substantial advances in the tasks of link prediction and knowledge graph completion, identifying geospatial relationships remains challenging, particularly due to the fact that spatial entities are represented with single-point geometries, and textual attributes are frequently missing. In this study, we present GTMiner, a novel framework capable of jointly modeling Geospatial and Textual information to construct a knowledge graph, by mining three useful spatial relationships from a geospatial database, in an end-to-end fashion. The system is divided into three components: (1) a Candidate Selection module, to efficiently select a small number of candidate pairs; (2) a Relation Prediction component to predict spatial relationships between the entities; (3) a KG Refinement procedure, to improve both coverage and correctness of a geospatial knowledge graph. We carry out experiments on four cities' geospatial databases, from publicly-available sources and compare with existing algorithms for link prediction and geospatial data integration. Finally, we conduct an ablation study to motivate our design choices and an efficiency analysis to show that the time required by GTMiner for training and inference is comparable, or even shorter, than existing solutions.

Formal topological description of three-dimensional topological entities based on k-dimensional pseudo-manifold

This paper clarifies the necessity of studying three-dimensional spatial data models, analyzes the research progress and existing problems of topological spatial relationship description methods, and proposes a complete and formal description framework of three-dimensional topological spatial relations based on point set topological theory and dimension expansion method. On this basis, the topological spatial relations existing in three-dimensional spatial objects are classified, and five fundamental topological spatial relations are defined. The mutual exclusivity and completeness of the minimal set of 3D topological space relations are proved. Based on point set topology and k-dimensional pseudo-manifold, this paper proposes the formal description of three-dimensional topological properties and spatial relations between spatial entities. The simplicial homology is applied to manifolds, the topological properties of three-dimensional space entities are revealed, and the formal description method of three-dimensional space entities is given through the geometric structural relationship between simplex and simplicial k-complex.

Exact Learning of Qualitative Constraint Networks from Membership Queries

A Qualitative Constraint Network (QCN) is a constraint graph representing problems under qualitative temporal or spatial relations. More formally, a QCN includes a set of entities and a list of qualitative constraints defining the possible scenarios between these entities. Qualitative constraints are expressed as disjunctions of binary relations capturing the (incomplete) knowledge between the involved entities. QCNs effectively represent various real-world applications, including scheduling and planning, configuration, and Geographic Information Systems (GIS). It is, however, challenging to elicit, from the user, the QCN representing a given problem. To overcome this difficulty in practice, we propose a new algorithm for learning, through membership queries, a QCN from a non-expert. Membership queries are asked to elicit temporal or spatial relationships between pairs of temporal or spatial entities. To improve the time performance of our learning algorithm, constraint propagation and ordering heuristics are enforced. The goal is to reduce the number of membership queries needed to reach the target QCN. We conducted several experiments on randomly generated temporal and spatial QCN instances to assess the practical effect of constraint propagation and ordering heuristics. The results of the experiments are encouraging and promising.

Events and objects are similar cognitive entities

Logico-semantic theories have long noted parallels between the linguistic representation of temporal entities (events) and spatial entities (objects): bounded (or telic) predicates such as fix a car resemble count nouns such as sandcastle because they are “atoms” that have well-defined boundaries, contain discrete minimal parts and cannot be divided arbitrarily. By contrast, unbounded (or atelic) phrases such as drive a car resemble mass nouns such as sand in that they are unspecified for atomic features. Here, we demonstrate for the first time the parallels in the perceptual-cognitive representation of events and objects even in entirely non-linguistic tasks. Specifically, after viewers form categories of bounded or unbounded events, they can extend the category to objects or substances respectively (Experiments 1 and 2). Furthermore, in a training study, people successfully learn event-to-object mappings that respect atomicity (i.e., grouping bounded events with objects and unbounded events with substances) but fail to acquire the opposite, atomicity-violating mappings (Experiment 3). Finally, viewers can spontaneously draw connections between events and objects without any prior training (Experiment 4). These striking similarities between the mental representation of events and objects have implications for current theories of event cognition, as well as the relationship between language and thought.

GeoGraphViz: Geographically constrained 3D force‐directed graph for knowledge graph visualization

AbstractKnowledge graphs are a key technique for linking and integrating cross‐domain data, concepts, tools, and knowledge to enable data‐driven analytics. As much of the world's data have become massive in size, visualizing graph entities and their interrelationships intuitively and interactively has become a crucial task for ingesting and better utilizing graph content to support semantic reasoning, discovering hidden knowledge discovering, and better scientific understanding of geophysical and social phenomena. Despite the fact that many such phenomena (e.g., disasters) have clear spatial footprints and geographic properties, their location information is considered only as a textual label in existing graph visualization tools, limiting their capability to reveal the geospatial distribution patterns of the graph nodes. In addition, most graph visualization techniques rely on 2D graph visualization, which constrains the dimensions of information that can be presented and lacks support for graph structure examination from multiple angles. To tackle the above challenges, we developed a novel 3D map‐based graph visualization algorithm to enable interactive exploration of graph content and patterns in a spatially explicit manner. The algorithm extends a 3D force directed graph by integrating a web map, an additional geolocational force, and a force balancing variable that allows for the dynamic adjustment of the 3D graph structure and layout. This mechanism helps create a balanced graph view between the semantic forces among the graph nodes and the attractive force from a geolocation to a graph node. Our solution offers a new perspective in visualizing and understanding spatial entities and events in a knowledge graph.

Introduction - Archaeology and the Natural Environment

Archaeological sites and monuments are defined as spatial entities and are therefore an intrinsic part of any environment, as humans perceive it. Landscape archaeology and concepts such as 'landscape biography' have taught us that our environment has developed over millennia of interactions between humans and nature. In co-evolutionary feedback-loops, human societies adapt to and change their environments and archaeological sites also reflect this principle.

Where are cities under pressure? – An indicator for measuring the impact of building changes on urban density

In the course of urbanisation, cities are becoming increasingly dense with high local variability. Worldwide, there is a high demand for living space, especially in larger cities, which leads to settlement pressure. It is a declared goal to further densify cities in order to avoid the consumption of ecologically valuable land. But cities that are too densely built also bring negative consequences. Therefore, it is important to understand and describe the densification processes of cities. To better describe settlement densification processes, we have developed a novel indicator that can be used to determine changes in the development density, which is of interest from a planning perspective. The indicator takes into account both the extent of the building change and the existing development. If an area is already intensively developed, a newly constructed building will result in a higher density than if that building were built in a previously undeveloped area. The indicator can also be used to measure deconstruction effects symmetrically, so that densification and de-densification processes can be analysed in parallel. In addition, various spatial entities such as cadastral parcels, grid cells or city districts can be used as reference areas for indicator calculation. In addition to an area-based analysis, we have also taken into account the number of storeys of each building, which allows us for more precise calculations. We tested the indicator in this study using the City of Hamburg as an example. Freely accessible building floor plans and parcels from 2015 and 2020, taken from the Authoritative Real Estate Cadastral Information System (ALKIS) of the City of Hamburg, served as the data basis. Furthermore, we used land use data to examine how various land use classes differ in terms of density development. The results show that extensive densification is often observed on previously built-up land, especially in the city centre, which indicates particularly high settlement pressure. Newly developed residential areas are also densely built up, but comparatively do not contribute to such high building density. On the one hand, this indicates an efficient and thus sparing use of the finite resource of land in the city. On the other hand, it points to increasingly dense development in the city centre.

Developing a Method to Detect Indoor Spatial Entities in Omnidirectional Images for Constructing IndoorGML Data

The continuous increase in interest in indoor space demands the development of LBSs (Location Based Services) for navigation, routing, and space query in these environments. Applications providing these services must use visualization and topology data together to provide relevant services to users. Applications have frequently used geometric object-based visualization models for indoor space LBSs. These have the advantage of being easy to use with IndoorGML, an indoor topology model, but have the disadvantage of high construction cost and heavy data. Correspondingly, the image-based visualization model drew attention as an alternative model. However, using such models requires identifying objects on the image to be used alongside IndoorGML, presenting a limitation because it is difficult to identify objects in the image pixels directly. To overcome the limitations of image-based visualization models and reconsider the usability of indoor space LBS using images, this study presents a method to automatically detect spatial objects required to construct IndoorGML from images using deep learning. This methodology aims to detect objects mainly used in indoor LBSs, which include door and stair objects portrayed in indoor omnidirectional images. This study proposes a detailed method of constructing a training dataset for indoor spatial object detection. Moreover, this research presents a method of refining the training dataset and directly acquiring omnidirectional image data to train it as an object detection model that can be used universally in various buildings while maintaining compliant accuracy.

Towards strategy of geoinformation systems and technologies use for territory management

GeoInformation (GI) Systems (GIS) and GI Technologies (GIT, together GIST) have been used for almost half a century, since the creation of Canada's first GIS in the 60s of the last century, to solve territory management problems. Over the past years, GISTs have reached their maturity, but still continue to develop, covering ever wider areas of use. Even the science of geoinformatics has emerged, in which GIST is used mainly as a toolkit or technology. As an example, geoinformatics in the same Canada is called geomatics and is a technology and/or technological science. At the same time, the expansion of the field of GIST use poses to researchers the question of methods and methodology. They are followed by issues of methods and methodology of geoinformatics not only as a technology, but also as a science. Moreover, these issues become more complicated with the expansion of the field of use. In the information industry, together with the field of use, the term "domain" or "context" is used. Thus, modern GIST usage manipulate a large number of interrelated terms and concepts that are often not clearly defined. The work is devoted to the classification of the main ones, which are influenced by the strategy selected. Spatial models of territory are used in the work. They are used in the study of both territorial systems of reality and individual spatial entities and phenomena of reality. Among spatial models, the main attention is paid to information spatial models, the most famous of which are GeoInformation Systems (GIS). GIS are inseparable from GIS tools - GeoInformation Technologies (GIT). The main results of the article were obtained using the so-called method of Conceptual Frameworks (CoFr) of Spatial Information Systems (SpIS). The CoFr method is applied to a special class of GIS - Atlas Geo-Information Systems (AGIS) of large territories (LT). The AGIS class includes Electronic Atlases (EA), Atlas Information Systems (AtIS), Cartographic Information Systems (CIS) and, in fact, GIS, if we are talking about LT. AGIS-LT is a hierarchical echeloned SpIS, for which the main terms and concepts of the article are applicable. These are such terms and concepts as "strategy" and "methodology" of GIS usage. GIS, in turn, use GIT, which are also classified using CoFr SpIS.

A spatiotemporal data model and an index structure for computational time geography

The availability of Spatiotemporal Big Data has provided a golden opportunity for time geographical studies that have long been constrained by the lack of individual-level data. However, how to store, manage, and query a huge number of time geographic entities effectively and efficiently with complex spatiotemporal characteristics and relationships poses a significant challenge to contemporary GIS platforms. In this article, a hierarchical compressed linear reference (CLR) model is proposed to transform network-constrained time geographic entities from three-dimensional (3D) (x, y, t) space into two-dimensional (2D) space. Accordingly, time geographic entities can be represented as 2D spatial entities and stored in a classical spatial database. The proposed CLR model supports a hierarchical linear reference system (LRS) including not only underlying a link-based LRS but also multiple higher-level route-based LRSs. In addition, an LRS-based spatiotemporal index structure is developed to index both time geographic entities and the corresponding hierarchical network. The results of computational experiments on large datasets of space–time paths and prisms show that the proposed hierarchical CLR model is effective at storing and managing time geographic entities in road networks. The developed index structure achieves satisfactory query performance in milliseconds on large datasets of time geographic entities.

The Sense of Belonging in the Context of Migration: Development and Trajectories Regarding Portuguese Migrants in Luxembourg.

The sense of belonging can be defined as a highly subjective and dynamic feeling of acceptance, inclusion, and connectedness to a specific contextual entity. Perceiving belongingness to others is positively related to psychological well-being, happiness, or higher self-esteem. The present contribution examined how the sense of belonging to spatial, social, and cultural entities evolves over the migration process of Portuguese first-generation migrants and their second-generation offspring residing in Luxembourg. The current study drew on the qualitative content analysis of ten semi-structured interviews, carried out with ten Portuguese migrant family dyads (one parent and one adult child per dyad). The results affirmed that the sense of belonging showed to be a complex and multi-faceted concept and highly shaped by specific contexts. The initially unfamiliar Luxembourgish context became a familiar setting and even a "home" over time. While the older generation presented noticeable belongingness to Luxembourg as a homeland, their affiliation to the Luxembourgish community and culture remained rather low. Simultaneously, they preserved a high connectedness to the Portuguese culture as well as to fellow Portuguese migrants living in the Grand Duchy. The younger generation expressed a much more pronounced attachment to Luxembourg, since they perceived belongingness to the Luxembourgish spatial, social, and (multi)cultural milieu. In addition to this, a certain affiliation to the Portuguese culture and language could be discerned. Although some factors, which might have contributed to this evolution, could be identified in the present study, one can assume that there are significantly more that have not been addressed yet.

Testing the spatial association of different types of polygons

In spatial statistics, we want to test hypotheses about the spatial association between two spatial patterns of geometric shapes. The spatial entities representing these geometric shapes are two-dimensional polygons, but researchers have often reduced these polygons to points to use point pattern methods. In this work, we developed techniques to test hypotheses related to polygon patterns that take advantage of all the information present in them. Our method is centered on a Monte Carlo toroidal shift approach to generate the reference null distribution. However, this approach comes with a caveat: the original polygons are broken into pieces. We were able to adapt the toroidal shift to work with polygon objects, eliminating this issue. In addition, extensions of the K12 function used for point patterns were proposed. We carried out a simulation study to demonstrate our method’s efficacy and to compare different test statistics. We illustrate the method by applying it to a real large data-set with different grass species.

Importance of the mapping unit on the land suitability assessment for agriculture

Land suitability assessment (LSA) provides geospatial information about growing crops where they are best suited and can play a crucial role in addressing contemporary challenges such as feeding 9 billion people by 2050, coping with climate change, and enabling sustainable production. Despite the known limitations of the current mapping units (conventional soil map units) used for LSA, alternative methods to objectively delineate mapping units, such as geographic object-based image analysis (GEOBIA) or geomorphons have never been tested for LSA. The objective of this work is to quantitatively assess the effects of different polygon-based mapping units on LSA for agriculture: 1) conventional soil map units; 2) units delineated semi-automatically using GEOBIA; 3) geomorphons. In addition, the three delineation procedures will be compared with the pixel-based LSA conducted as a benchmark. LSA is conducted within the framework of the existing Romanian rating methodology for land suitability, which was developed based on FAO guidelines for land evaluation. We use georeferenced soil profiles with field-measured soil properties and digital terrain models to digitally map 17 eco-pedological indicators (e.g. soil pH, soil texture, soil porosity, gleization, carbonate content, humus content). Based on several lookup tables, these maps are transformed into digital maps with suitability ratings for 14 crops, 7 fruit trees, and 2 land-use types, ranging from 0 (not suitable) to 1 (maximum suitability). The product (multiplication) of the 17 maps with LSA ratings is the final suitability map for each crop and land use. Overall, the best maps were obtained when the LSA was conducted using the GEOBIA units (similar accuracy to the pixel-based approach), whereas geomorphons and conventional soil map units resulted in much poorer maps. GEOBIA is much more suitable for LSA than the conventional soil map units and geomorphons, showing a much higher accuracy and internal homogeneity. Another conclusion is that the LSA based on conventional soil map units recorded the poorest accuracies by far. Our results show that the GEOBIA-based units or pixel-based approach are the right choices when conducting LSA, however, if the map user (eg. farmer, land manager) needs delineated units as semi-permanent and stable regions to manage them as stable spatial entities then GEOBIA technique should be used for LSA.