Land Use Change Model Research Articles

Can regionalization enhance the performance of land-use change models in rapidly urbanizing areas?

The performance of Land Use Change (LUC) models is influenced by the regional spatial characteristics that trigger the changes. However, the literature on LUC models generally reports validation results for entire regions without considering subregions that differ significantly in their LUC drivers. This research explores how the LUC driving forces differ among subregions and whether regionalization can improve the performance of LUC models in areas undergoing rapid urbanization. We analyzed the Geomod, Cellular Automata-Markov, and Land Change Modeler models across rural and urbanized subregions on the western edge of Mexico City. Regionalization significantly enhanced the overall accuracy of the models and the concordance of spatial patterns with the reference data in rural regions but was of limited benefit in urbanized regions. This shows the need to consider regionalized modeling to improve the performance of LUC models when there are noticeable differences in LUC drivers between subregions. These findings will enhance the usefulness of LUC models for urban planning and land management policies, promoting more precise and effective decision-making.

Urban growth and waste services: exploring scenarios in an inter-municipal cooperative framework in Sogamoso city, Colombia

ABSTRACT The spatial dimensions of city expansion in Latin America reveal the emergence of new urban agglomeration and metropolization phenomena. In this context, the goal is to implement a predictive model of land use change between 1976 and 2047 in the city of Sogamoso and neighboring municipalities in Colombia and to apply its results to a proposal for inter-municipal cooperation in the collection of solid household waste. The methodology used is based on a spatially explicit models using the CA_Markov model in Idrisi Selva software that incorporates Markov chains, multi-criteria evaluation, and cellular automata to identify future areas of urban expansion. The model forecasts urban growth from 2,420 ha in 2021 to 5,278 ha in 2047. The spatial distribution of the new urban areas implies collection routes that are ever farther away from the main urban centers. There is thus a vital need for new regional planning practices (such as inter-municipal cooperation).

Digitalization and Spatial Simulation in Urban Management: Land-Use Change Model for Industrial Heritage Conservation

Digitalization and Spatial Simulation in Urban Management: Land-Use Change Model for Industrial Heritage Conservation

Modeling of land use and land cover changes using google earth engine and machine learning approach: implications for landscape management

A precise and up-to-date Land Use and Land Cover (LULC) valuation serves as the fundamental basis for efficient land management. Google Earth Engine (GEE), with its numerous machine learning algorithms, is now the most advanced open-source global platform for rapid and accurate LULC classification. Thus, this study explores the dynamics of the LULC changes between 1993 and 2023 using Landsat imagery and the machine learning algorithms in the Google Earth Engine (GEE) platform. Focus group discussion and key informant interviews were also used to get further data regarding LULC dynamics. Support Vector Machine (SVM), Random Forest (RF), and Classification and Regression Trees (CART) were demonstrated for LULC classification. Six LULC types (agricultural land, grazingland, shrubland, built-up area, forest and bareland) were identified and mapped for 1993, 2003, 2013, and 2023. The overall accuracy and kappa coefficient demonstrated that the RF using images comprising auxiliary variables (spectral indices and topographic data) performed better than SVM and CART. Despite being the most common type of LULC, agricultural land shows a trend of shrinking during the study period. The built-up area and bareland exhibits a trend of progressive expansion. The amount of forest and shrubland has decreased over the last 20 years, whereas grazinglands have exhibited expanding trends. Population growth, agricultural land expansion, fuelwood collection, charcoal production, built-up areas expansion, illegal settlement and intervention are among causes of LULC shifts. This study provides reliable information about the patterns of LULC in the Robit watershed, which can be used to develop frameworks for watershed management and sustainability.

Broadening the horizon in land use change modelling: Normative scenarios for nature positive futures in Switzerland

Within scenario-based research of social-ecological systems, there has been a growing recognition of the importance of normative scenarios that define positive outcomes for both nature and society. While several frameworks exist to guide the co-creation of normative scenario narratives, examples of operationalizing these narratives in quantitative simulation modelling are still limited. To address this gap, this paper presents an example of how aspects of normative scenarios can be realized within a spatial model of land use and land cover change. This is achieved through a combination of data-driven approaches to encapsulate scenario-specific differences in local and global scale phenomena, as well as iterative expert elicitation to quantify descriptive trends from narratives. This approach is demonstrated with a case study simulating five scenarios of landscape change (three normative and two exploratory) in Switzerland between 2020 and 2060. The resulting maps of future land use and land cover exhibited distinct variations between the scenarios, notably with regard to the prevalence of areas of heterogeneous semi-natural land, such as alpine pastures and grassland, often considered culturally emblematic of Switzerland. While the simulation results were generally consistent with the outcomes expressed in the scenario narratives, following a process of expert feedback, we reflect that there are clear challenges in leveraging such results to elicit further discussions as to the desirability and plausibility of future scenarios. Specifically, the need to summarize spatial simulations in a manner that is easily interpretable and encourages consideration of the broader patterns of change rather than focusing on fine-scale details.

Selected Driver Variables for the Simulation of Land-Use and Land-Cover Change for the Republic of Djibouti: A Study from Semi-Arid Region

Abstract. This study aims to integrate driver variables with a land use change model (LCM) to explore their impact on the natural environment within the context of land-use changes in the Republic of Djibouti, considering possible Business-as-usual scenarios. Secondary data from 1990 and 2012 on land use land cover (LULC) were analyzed, with a 2022 map generated by adopting the same method of secondary data used (random forest classification in Google Earth Engine (GEE)) for validation. Eight key driver variables were utilized to model plausible future land cover (2035) for Djibouti. Statistical outputs and change maps from the LCM were compared to gauge historical change estimates and simulated scenarios. Analysis from 1990 to 2022 highlights significant land use and cover changes spurred by urbanization, environmental factors, and economic development. Barren land and bushland dominated, while built-up areas and water bodies expanded notably. Urbanization, agriculture, and climate change contributed to vegetation degradation, with declines in mangroves and increases in built-up areas. Water bodies also expanded during this period. Projections from the 2035 LULC map anticipate further urban expansion, underscoring the need for sustainable land management practices. In conclusion, comprehensive land-use planning, interdisciplinary approaches, and stakeholder engagement are deemed critical for addressing Djibouti's socio-economic and environmental challenges and steering towards a sustainable future. These simulated results offer valuable insights for regional governments to frame strategic policies and assess management actions for resource utilization amidst urbanization and population growth trends.

Greenbelt's impact on urban forest habitat: Insights from morphological analysis in quantifying conservation effects

Urban forests play a crucial role in preserving biodiversity in highly urbanized areas by providing habitats for various species and promoting ecological connectivity. However, managing urban forests presents challenges such as low habitat quality and fragmentation risks. To address these challenges, protected areas serve as a fundamental policy tool. This study analyzes the effect of Greenbelt areas, protected spaces within urban settings, on future land-use and habitat changes. Through morphological analysis of forest areas, this study attempts to analyze the current status of habitats in cities affected by fragmentation and derive the effects of green belts, thereby presenting the direction of biodiversity conservation for city habitats. The Slope, Land-use, Excluded area, Urban, Transportation, Hillshade (SLEUTH) land-use change model was used to predict future land-use changes by applying a scenario that adjusts weights within the Greenbelt areas. In addition, the Morphological Spatial Pattern Analysis (MSPA) Tool was used to derive changes by habitat type, which aided in deriving the implications of the Greenbelt regulation on the habitat. Our findings indicate that Greenbelt regulation effectively prevents the reduction of critical Core areas and minimizes damage to Islets, small patches. Comparisons between Core and Edge areas reveal that Edge areas are more vulnerable to development, and the impact of regulating the Greenbelt area is significant. In addition, Greenbelt regulations can help prevent habitat fragmentation by reducing Perforation, which creates cavities within Core areas. This study quantitatively demonstrated that the Greenbelt can play an important role not only in preventing urban expansion but playing an important role in habitat conservation in cities.

Application of land-use change model for supporting land-use planning in An Bien district, Kien Giang province, Vietnam

Forecasting land-use change is a critical step in assisting land-use planning, especially for coastal areas that are influenced by many economic and natural environmental factors. This study aims to estimate land use changes based for An Bien district, Mekong Delta based on multi-criteria assessment of factors affecting land use change. The model has been calibrated with the 2015 land-use map and verified with the 2020 map. The results of land-use change analysis in An Bien district were carried out according two scenarios: (i) forecasting the area under present conditions and (ii) forecasting land use area in the case of economic benefits and suitable environmental conditions for aquaculture production. Scenario 1 showed that the results will not change much compared to the current situation, in which rice-shrimp plays a leading role locality, however Scenario 2 showed that shrimp area will rapidly rise. In general, the simulation results allow users to explore future planning possibilities and serve as a basis for selecting appropriate land-use types.

A cellular automata model coupled with partitioning CNN-LSTM and PLUS models for urban land change simulation

Urbanisation is a key aspect of land use change (LUC), and accurately modelling of urban LUC is crucial for sustainable development. Cellular automata (CA) are widely used in LUC research. However, previous studies have overlooked the significant temporal dependence and spatial heterogeneity associated with LUC. To address these gaps, this study proposes a novel model called KCLP-CA, which integrates k-means, a convolutional neural network (CNN), a long and short-term memory neural network (LSTM), and the popular patch-generation land use model (PLUS). Initially, k-means and CNN are utilised to address spatial heterogeneity, while LSTM tackles temporal dependence. The LSTM and land expansion analysis strategy (LEAS) models of PLUS are employed to obtain land use conversion probability maps. Finally, a simulation of land use dynamic change was conducted using a linear weighted fusion conversion probability map that accounts for random factors. To validate the KCLP-CA model, land use data collected from Hangzhou between 1995 and 2000 were employed. The results showed that the KCLP-CA model outperformed traditional methods, including artificial neural networks and random forest model, with the figure of merit (FoM) index increasing from 2.12% to 4.19%. Random forest analysis of drivers impacting LUC revealed that distance to water and road network density exerted the greatest influence on urban land development in Hangzhou. Incorporation of various policy planning factors affecting urban development yielded simulation results aligning more closely with reality, resulting in a FoM index increase of 1.64–1.76%. In summary, the model developed in this study combines the strengths of two sub models to deliver an accurate and effective simulation of future land use.

Driving model of land use change on the evolution of carbon stock: a case study of Chongqing, China.

Terrestrialecosystems are significant carbon sinks and are crucial for understanding the regional and global carbon cycles, energy flow, and climate change. As land use change is a significant process affecting ecosystem carbon stocks and striving for land degradation neutrality (LDN), studying it is essential for comprehending the evolution of regional carbon sink functions and achieving sustainable development goals. The drastically diverse land use patterns in each of the study area's regions resulted in significant differences in carbon stock. This study explores the evolution traits of carbon stocks based on land use data and their driving mechanisms in Chongqing during the past 30 years by using spatial analysis, the InVEST model, and geographic probes. The results demonstrate that from 1990 to 2020, land degradation in Chongqing was made worse by the demand for land for construction land, but the strategy of converting cropland back to forests raised the carbon stock of forest land. The overall result is a decrease in total carbon stocks of 5.1078 Tg or 1.5%. The main pathway for carbon loss pathway in the evolution of carbon stock is the conversion of cropland to construction land, and the primary carbon compensation pathway is the conversion of grassland and cropland to forest land, with a spatial distribution characterized by "higher in the whole area and obvious local differences." The land use intensity index has the most significant influence on the evolution of carbon stock. Moreover, the interaction of pairwise factors played a more important role in affecting the evolution of carbon stocks than did each factor individually. The case study in this paper shows that land use change is a significant driving mechanism for the evolution of carbon stock, and the development of a driving model theory is appropriate for deciphering the trajectory of carbon stock evolution and offering research suggestions for other regions.

Automatic image segmentation model for indirect land use change with deep convolutional neural network

Automatic image segmentation model for indirect land use change with deep convolutional neural network

Mapping and Modelling of Land Use Change in Nepal

Land use changes are a common occurrence resulting from different factors in Nepal. There is a need for better understanding of the temporal and spatial change dynamics of land uses as well as their future projections. This paper assesses the land use change and its impact on carbon stocks and provides future scenarios of land use changes and their impacts on carbon stocks. The methods involved classification of medium resolution satellite image using Object Based Image Analysis (OBIA) techniques, analysis of the change across three time frames,modelling the future land use, and assessment of the impacts on carbon stocks. OBIA was used for the image segmentation and image classification for the assessment of land uses present in two different periods i.e. 2010 and 1990. The classification of the images was carried out using the spectral characteristics of the Landsat images. Image indices, i.e., Vegetation Index were used for land use classification.
 The land use changes over time and space was analysed using Geographical Information System (GIS) overlaying techniques. A future scenario model was prepared using Land Change Modeler of IDRISI software. The study has focused on the land use changes and their future projection based on the potential transition of the land uses. The comparison of land use maps from 1990 to 2010 shows that the different pattern of annual land use changes in different districts such as forest areas are increasing at annual rate of 0.3 per cent in Rasuwa, 2.3 percent in Gorkha and 0.6 per cent in Chitwan. Similarly, the agricultural land is decreasing at the annual rate of 0.2 per cent in Rasuwa, 2.4 per cent in Gorkha and 0.8 per cent in Chitwan. Likewise, the future projection of land uses in all the three sites was made using the previous land use pattern and driving forces. There was a 0.1 per cent change annually in forest coveragein both Gorkha and Rasuwa whereas Chitwan showed higher rate of forest cover change of 0.2 per cent annually.

Assessment of ecological security risk in rocky desertification area based on land-use change model

Social and economic development, production and living needs, other human behavior requirements and various ecological engineering construction are considered as the main influencing factors of land spatial pattern changes and land ecological security risks in rocky desertification areas. From the perspective of land-use spatial change, the impact of human activities and resource utilization on land ecological security risks can be more intuitively analyzed. Based on the land-use change survey data of Phase 4 of Honghe Prefecture, China, the study measures the characteristics of land-use transfer flow and land-use dynamic change, analyzes the main influencing factors and comprehensively constructs the land ecological security risk index model, on which the land ecological security risk is classified. The results showed that the land ecological security risk degree in the three stages of the study area from 2014 to 2020 was 2016–2018 > 2018–2020 > 2014–2016, and had obvious regional characteristics. Among many influencing factors, the different activities/projects (positive or negative) types, time and intensity affect the land-use transfer flow and dynamic change characteristics, resulting in corresponding changes in the risk of land-use ecological security. Among them, man-made destruction and construction activities have the most significant negative effects on land ecological status. The empirical analysis proved that building an ecological security risk index based on the direction, area, and rate of spatial changes in land-use can accurately reflect the degree of desertification risk in rocky desertification areas. The study results can provide important information on the impact of regional development and the environmental governance policies of the land-use change in rocky desertification areas.

ANALISIS SPASIAL PERUBAHAN PENGGUNAAN LAHAN DI KABUPATEN PURBALINGGA

The rapid development of the Purbalingga Regency region, supported by the establishment of increasingly complex regulations and significant population growth, has significantly impacted the demand for facilities and infrastructure development; this aims to realize development plans and meet the increasing needs of society. Changes in land use are an important factor that needs to be analyzed and planned carefully. This research aims to understand changes in land use in Purbalingga Regency. The method used involves interpreting Landsat images with predictive modeling of land use changes using Annual Neural Network and Cellular Automata (ANN-CA) to describe land use changes between 2009 and 2022 using data techniques observation. The research results show that the most dominant land uses in Purbalingga Regency include gardens, rice fields, forests, and settlements. The most striking change in land use is the transformation from forest and agricultural land to residential land. The residential land area 2009 was around 6,294.48 hectares, which increased significantly to around 15,183.88 hectares in 2022.

Exploring potential futures: Evaluating the influence of deep uncertainties in urban planning through scenario planning: A case study in Rome, Italy

Cities play a critical role in developing adaptable strategies to address the challenges posed by climate change. However, the inherent complexity of urban environments and their uncertain future conditions necessitate exploring innovative approaches and tools to assist the current planning practices. This paper presents a workflow rooted in model-based scenario planning for long-term adaptation planning given uncertain futures. To demonstrate the workflow’s effectiveness, a pertinent case study was conducted in a flood-prone area of Rome. The study employed a land-use change model to examine potential urban growth patterns, considering the uncertain implementation of new poles of attraction. This interdisciplinary study constitutes an initial stride toward implementing uncertainty within urban planning frameworks. Future prospects encompass the integration of multiple models for cross-scale analysis, embracing further critical environmental and social aspects. This research contributes to advancing urban resilience strategies. It enhances the understanding of adapting to an uncertain future in the face of climate change, as urban areas must embrace comprehensive planning to ensure flexible adaptation when faced with climate-driven uncertainties in long-term planning. In conclusion, the study underscores that embracing uncertainty is a challenge and a pivotal opportunity to shape resilient and adaptable urban futures.

Simulating Seoul's greenbelt policy with a machine learning-based land-use change model

This study builds a machine-learning-based land-use change (ML-LUC) model to analyze the effect of green belt (GB) regulation in the Seoul metropolitan area (SMA) and predict the spatially explicit development potential of the land within the GB under the assumption of a no-GB policy scenario. It stands out for its ML-LUC application to simulate counterfactual planning for urban land use regulation. After comparing the predictive power of extreme gradient boosting (XGB), random forest (RF), and artificial neural network (ANN), the ML-LUC model utilizes the XGB algorithm due to its outperformance. Three scenarios based on SMA's land market demand were simulated to estimate the potential population and employment within the GB under the no-GB policy: high, moderate, and low land market demand. The results suggest 0.6 to 1.5 million residents, 0.2 to 0.5 million manufacturing jobs, and 0.4 to 1.0 million service sector jobs could have been located within the GB, accounting for 3 % to 6 % of total population and 5 % to 13 % of all employment in SMA. The findings imply the GB regulation prevents population and employment from locating within the GB, pushing them to central Seoul or suburbs beyond the GB under a closed-city assumption.

Planning to ‘Hear the Farmer’s Voice’: an Agent-Based Modelling Approach to Agricultural Land Use Planning

Agricultural land use is influenced not only by multiple aspects of biophysical and socio-economic processes, but also the cumulative impacts of individual farmer decisions. Farmers’ activities and decisions at farm scale shape land use and water utilisation at regional scale, yet land use planning processes do not take into account farmers’ knowledge and decision-making processes as they respond to, and in turn shape, change. Farmers’ voices are missing in the planning system. In this paper, we address the complexity of agricultural land use planning and examine the possibility of agricultural land use planning from the bottom-up via simulation to integrate environmental, economic and human factors that influence land use change. We present an innovative approach to model the interactions between government policy, market signals, and farmers’ land use decisions, and how the accumulated effects of these individual decisions change agricultural land use patterns at regional scale, using spatial and temporal agent-based modeling. A multi-stage mixed method spatial agent-based modeling (ABM) approach, aligned with the Geodesign framework, can incorporate local knowledge and decision-making into models of regional land use change. To illustrate the new approach, we examine the impact of milk market price on changes in land use in Tasmania, Australia. This approach brings together local knowledge with scientific, planning, and policy knowledge to generate dynamic scenarios for informed agricultural land-use planning decisions.

Long Short-Term Memory and Attention Models for Simulating Urban Densification

This paper introduces a novel cellular automata model that combines Long Short-Term Memory, Attention, and Neural Network models to capture spatio-temporal Land Use Change (LUC) behaviors while addressing the challenge of imbalanced datasets. The proposed method is developed and validated using data from Belgium, defined as three (100x100) m raster-based built-up maps for 2000, 2010, and 2020. The model is trained and validated using data from 2000 to 2010, and its effectiveness is tested using data from 2010 to 2020. The key contribution of our approach lies in its ability to tackle long-term temporal dependency and class imbalance problems in LUC science. Our proposed method significantly enhances the performance of spatio-temporal LUC simulation. Additionally, we adopt a data splitting strategy that takes into account the different transitions between classes, improving the accuracy of the model predictions of minority class. The obtained results demonstrate the efficiency of the proposed model in capturing complex spatio-temporal dynamics and reducing the impact of imbalanced datasets surpassing existing methods. The implications of our study extend beyond LUC modeling, as the proposed approach can be applied to a wide range of applications where machine learning is used to model complex environmental and geographical phenomena.

Enhanced carbon storage in semi-arid soils through termite activity

Termites are keystone species in natural ecosystems and their role in the C cycle is potentially substantial but poorly understood. Large (20–40 m) mounds (heuweltjies) of the harvester termite Microhodotermes viator occupy up to a quarter of the semi-arid west coast region of South Africa but their C storage potential is unknown. This study determined the organic and inorganic C fractions, C stocks, and their correlation with each other, depth, and biogenic features in these mounds. Trenches (30–60 m) were excavated through 3 mounds: Buffels River (m.a.p < 100 mm), Klawer (m.a.p 100–200 mm) and Piketberg (m.a.p 300–400 mm) and grid sampled. Mound soils had significantly higher soil organic carbon (SOC) and inorganic carbon (SIC) than surrounding soils. Total C was strongly correlated (ρ > 0.9; p < 0.001) with SIC in the arid mounds and SOC (ρ > 0.75; p < 0.001) in the higher rainfall mound. There was no consistent relationship between SOC and SIC distributions throughout the mounds, which is likely related to solubility-linked translocations of carbonates. For all mounds, SOC was highest in topsoils with a second clear peak in subsoils (>1 m) that was associated with biogenic features, termite channels and burrows. Subsoils contributed substantially (36–41 %) to the total C stock. Total C stocks for the intermediate rainfall mound (Klawer) were estimated at 14.6 tons per mound, with 1.1 tons SOC. In this region, mounds occupy 27 % of the total area but contribute 44 % of the total SOC stock to a depth of 80 cm. This highlights the disproportionate contribution termite mounds make to carbon stocks of these semi-arid environments and demonstrates the importance of deep (<1 m) soil carbon for C modelling. Termite activity needs to be recognized as a major contributor to C stock variability both laterally and at depth and accounted for in land-use change (CO2-LULUCF) models.

Modeling of land use and cover changes (LUCC) in Deli Serdang Regency, North Sumatra Province

Land use/cover (LUC) is a substantial factor in land management and can influence policy in an area. LUC has the potential to change due to physical, economic, and social aspects. This study aims to analyze the spatialland use and cover changes (LUCC) in Deli Serdang Regency for the 2010 to 2020 period and predict LUC in 2030. The analysis was run by applying the Cellular Automata-Markov Chain method. The driving factors used in this modeling are the distance to the road, the distance to the river, population density, the distance to the district capital, and the distance to Medan city. The results showed that Kappa for image classification was 0.86. The dominant type of LUC in Deli Serdang Regency is a plantation, with a total area of more than 45%, followed by paddy fields, dryland agriculture, forests, and settlements/built-up areas. LUCC model validation obtained a kappa value of 0.89 (very good category) and can be applicated for predicting land use change models in 2030. By 2030, the settlements/built-up area and dryland agriculture will increase significantly, which 21,060 ha and 4,587 ha, respectively, while forests, plantations, and paddy fields will decrease significantly by around 9,266 ha, respectively, respectively 8,306 ha and 7,806 ha.