Land is one of the indispensable basic resources in human life and production. Land use is a process in which people use and renovate land by selecting different natural attributes to meet people’s needs for life and society. The land carries human activities in different time and space, thus creating an extremely complex and varied land use pattern. The pattern of land use, in turn, will have a positive or negative impact on human activities.
The land-use/land-cover (LULC) change may affect climate, ecosystem processes and biodiversity . Therefore, understanding and modeling of LULC change is of great significance to environmental protection and land use planning . A full understanding of the interaction between the driving forces of LULC change and a complete simulation is a prerequisite for accurately predicting future land use changes.
The driving force problem has always been a dominant aspect in the land use study . There are many factors that can drive changes in land use, including natural, economic, political, social, cultural, etc. . All the above factors are often divided into two categories: natural factors and socio-economic factors by the essential attributes [5,6]. Natural factors mainly include local climate, geological features and vegetation. The social and economic factors mainly include population change, economic development, policy and planning . Identifying the primary causes and estimating the processes are crucial for land use planning, utilization of regional resources and environment management .
Previous studies on the driving factors were mostly concentrated in the field of topographic , climate changes , population growth , urbanization [12,13], changes in the direction of economic development , borderline effect  and other fields. However, there are few studies associating natural and socio-economic factors. On the other hand, the common points of the above studies are large-scale, liking watershed or city, which emphasizes a macro perspective of the land use change. It lacks small-scale research, such as county scale.
To understand the process of LULC changes, LULC-change models are inevitable . In recent years, many models have been used to simulate and predict land use change, such as conversion of land use and its effects on small regional extent (CLUE-S) model [17,18], which is one of the most widely used land-use models, Markov model , multiagent systems (MAS) model [20,21], cellular automata (CA) model [22,23] and other models [24,25]. In most cases, these models showed relatively good results. However, in fact, land use changes are always caused by complex factors. The neighborhood scale, conversion rules, and simulation mechanisms would influence the CA’s accuracy of the simulation . Markov models are susceptible to location factors and decision-making processes. These models are all limited to theories and methods . So, there is a need to integrate these models to improve the effect of simulating land use, like CA-Markov model.
Landscape prediction is the basis of urban planning. The simulation results can help to understand the current urban development trends and related land use issues and to examine the consequences of urban evolution [27,28]. At present, there are some studies on the prediction of land use types at the county level [29,30]. Cellular automata-Markov model is an important prediction model, but there is no study on the affected factors at the county level. To explore the impact of land use and socio-economic factors on the prediction of CA-Markov model on county scale, the paper uses the CA-Markov model to simulate the land use of Anren County, China, and analyzes the prediction accuracy of the model.
2.1 Study area
Anren County is located in the south-east of Hunan Province (113° 05′ to 113° 36′ E, 26° 17′ to 26° 50′ N), China. It consists of 17 rural communities, such as Yongle River Town, Anping Town, etc. (Table 1). Yongle River Town is the political, economic and cultural center of Anren County. The total area coverage of the study area is 1,461 km2, and the elevation is descending from south-east to north-west. The annual precipitation average is 1,430 mm, and the annual temperature average is 17.7°C. Anren County has convenient transportation adjacent to eight counties and cities. In 2013, Anren County had a total population of 4,47,800 and per capita income is 5,448 yuan.
Population and economic situation of towns and townships in Anren County in 2013
|Towns and townships||Population (person)||Population density (person/km2)||Per capita income (yuan/person)|
|Yongle River Town||1,39,178||397||9,530|
2.2 Cellular automata-Markov chain prediction
Cellular automata was originally proposed by Ulam and Neumann in the 1940s, aiming to provide a formal framework for studying the behavior of complex, self-reproducible systems . It is a dynamic system that evolves in discrete space-time dimensions and has the ability to simulate the spatiotemporal evolution of complex systems. The cellular automaton consists of the following four parts: cell, lattice, neighborhoods, and rule. It should also include cell states and discrete time. As a kind of grid dynamics model that can be used to process a large number of variables, the core feature of the CA model is the generation of complex global patterns by predefining a set of simple transformation rules [32,33]. Cellular automata is a suitable method to simulate and predict the temporal and spatial evolution of complex geographical processes [34,35]. It is widely used in physical science, geography, ecology and mathematics .
The Markov chain model is a spatial probability model based on the grid data using the Markov chain, which takes advantage of the evolution from t − 1 to t to the probability of future t + 1 land use change; it is a suitable tool to simulate the future LULC dynamics [1,37]. It can show the direction and extent of future land use change in the region . The model is widely applied to the prediction of geographical features, which lacks after-effect events; it has also been an important prediction technique for the study of geography .
The Markov chain and cellular automata analysis model is a hybrid of the CA and Markov models ; it is suitable for the detection and simulation of land use change [1,40]. In the CA-Markov model, the Markov chain process controls temporal dynamics among the land use types through the transition matrices [1,40]. Cellular automata model controls spatial dynamics through the local rules considering either neighborhood configuration and transition probabilities [37,41,42]. Geographic information system (GIS) and remotely sensed data can be used to define initial conditions, parameterize the CA-Markov model, calculate transition probabilities, and determine the neighborhood rules .
In this study, TerrSet software is used to simulate the land use of the CA-Markov area. One of the functional modules of TerrSet is the land cover change simulation model, which provides the user with the CA-Markov model. The most characteristic is that CA-Markov model can define the transformation rules between land use types through multicriteria evaluation system and multiobjective decision support system . The TerrSet CA-Markov model strengthens the simulation ability of the spatial pattern based on the accurate prediction of the number structure of land use in the future with Markov chain. Therefore, we can effectively predict the quantity and spatial distribution of land use structure. The CA-Markov model has greater accuracy advantage than GIS-based technology in the simulation of land use change .
2.3 Analysis process
This study firstly used ENVI 5.0  for remote sensing images of Anren County in 1996, 2006 and 2016, included geometric correction, atmospheric correction, image cropping, etc. The above images belonged to Landsat Thematic Mapper (TM) or Operational Land Imager (OLI) and downloaded from Geospatial Data Cloud (http://www.gscloud.cn/). Then, land use raster data maps were extracted by the supervised classification method. The land use map was analyzed by ArcGIS [47,48], and the area and proportion of land use types in Anren County were calculated. The landscape metrics were reported by Fragstates . A total of seven representative landscape metrics belonging to class level and landscape level were selected in the study: number of patches (NP), landscape shape index (LSI), largest patch index (LPI), patch density (PD), patches percentage of landscape (PLAND), Shannon’s evenness index (SHEI) and perimeter-area fractal dimension (PAFRAC). Then, we simulated the land use in 2016 by TerrSet software with CA-Markov model based on the 1996 and 2006 land use raster maps reported in ref.  and compared with the actual land use in 2016 to get the kappa coefficient. Finally, the factors affecting land use simulation were analyzed by drawing scatter plots and correlation analysis through R version 3.5.1  to analyze the correlation between kappa coefficient, landscape metrics and socio-economic data. The “ggplot2” and “Hmisc” of R packages were used to make scatter plots and to calculate Pearson correlation coefficient (Figure 1). The core data, including the land use map of Anren in 1996, 2006, 2016 and simulated land use in 2016, were reposited in Dryad (https://datadryad.org) and could be accessed by doi: 10.5061/dryad.s1rn8pk4g.
3 Result and analysis
3.1 Distribution of land use and the change
The land use type was classified into built-up land, cultivated land, barren land, forest land and water (Table 2). Thirty years from 1996 to 2006 and 2016, cultivated land and forest land have always been the main types of land use in Anren County, consistently accounting for more than 90% of the total area. In addition, significantly changed area was observed in forest land (from 712.73 km2 to 844.68 km2 and 693.46 km2) compared with cultivated land (from 664.14 km2 to 576.47 km2 and 677.87 km2). In contrast, the area of barren land has always been the smallest and a tendency toward lower was seen from 1996 to 2016 (from 3.58 km2 to 3.02 km2 and 2.85 km2). The area of water was unstable, it was reduced by 55% from 1996 to 2016 (from 54.67 km2 to 18.82 km2 and 23.53 km2). Built-up land remained at a relatively low level, although a trend was witnessed for the increase in 30 years (from 27.74 km2 to 19.87 km2 and 65.15 km2).
Area statistics of different land types in 1996, 2006 and 2016 (km2)
|year Land use type||1996||2006||2016|
3.2 Distribution of landscape and the change
As shown in Table 3, the values of NP in Yongle River Town were 9,455, 10,273 and 7,582 in 1996, 2006 and 2016, far more than other townships in Anren County. During the period from 1996 to 2016, the NP in Anren County decreased (from 42,830 to 29,008) and most of the towns also decreased. It reflects that the reduction of the fragmentation of the landscape in Anren County. Largest patch index is the percentage of the total landscape area occupied by the largest patch and a representation of the degree of landscape dominance. The largest patch of Haoshan Township was accounting for a large proportion compared with other townships (from 91.01% to 95.91% and 85.04%). The LPI values in Anren County and other most townships were declined from 1996 to 2016. In terms of LSI, the patches’ shape of Yongle River Town and Pingshang Township is more complicated than other townships. The LSI values of various towns in Anren County were increased during the study period, showing that the overall landscape shape of Anren County is very complicated. In PAFRAC, the values of towns in 2016 were all below 1.50. It indicates that the patches’ shape of Anren County is getting simpler. Shannon’s evenness index can compare the diversity of landscapes in different periods and describe the distribution uniformity of various landscape types. The overall SHEI of Anren County remained unchanged, and the SHEI of each township increased or decreased. It means the landscape heterogeneity characteristics were obvious.
Landscape metrics analysis on county scale
|Yongle River Town||7,582||25.46||52.22||1.45||0.62||10,273||21.34||61.51||1.52||0.55||9,455||48.55||47.62||1.48||0.60|
NP: number of patches; LPI: largest patch index; LSI: landscape shape index; PAFRAC: perimeter-area fractal dimension; SHEI: Shannon’s evenness index.
Landscape metrics were analyzed at the class level in 2016 (Table 4). In terms of the NP, the number of built-up land patches was the largest (10,484) and barren land was the least (840). In the PLAND, the proportions of forest land and cultivated land were accounting for 47.31% and 46%, respectively. In the PD, the density of the built-up land was the largest (7.17), indicating that the degree of fragmentation is high, which is mainly caused by human activities. The density of patch in barren land was the smallest (0.57). The LPI values of cultivated land and forest land were higher than other land use types, indicating that the cultivated land and forest land were dominant in Anren County. The size of the PAFRAC value reflects the complexity of the shape of the landscape. The value of water was the smallest (1.38), which shows that the shape of water is relatively simple and is greatly disturbed by human beings.
Landscape metrics analysis at the class level in 2016
|Land use type||NP||PLAND||PD||LPI||PAFRAC|
NP: number of patches; PLAND: patches percentage of landscape; PD: patch density; LPI: largest patch index; PAFRAC: perimeter – area fractal dimension.
3.3 Land use predicted by CA-Markov model
In the simulated land use types of Anren County in 2016 (Figure 2), the area of forest land was the largest with 857.25 km2, occupied 58.60% of the total land use area of Anren County. Followed by cultivated land, with cultivated land area of 568.71 km2, accounting for 38.87% of the total area. The area of the built-up land was 20.50 km2 and the proportion was 1.40%. The area of water was 12.96 km2, accounting for 0.89%. The area of barren land was the smallest, only 3.55 km2 (Figure 2a).
By comparing the land use of Anren County in 2016 and the simulated land use in the same year, the overall kappa coefficient is 0.8048. It shows that the simulation land use of Anren County in 2016 is highly similar to the land use of Anren County in 2016. There were six townships with kappa coefficient above 0.8, including Anping Town, Guanwang Town, Haoshan Township, Pailou Township, Yangnao Township and Yongle River Town. Among them, Haoshan Township has the highest kappa coefficient of 0.8797. In others, the kappa coefficients of Longhai Town and Pingbei Township are relatively low, 0.6968 and 0.6510, respectively.
3.4 Relationship between kappa coefficient and landscape, population, income, respectively
The correlation between all calculated landscape indices and kappa coefficients is quite different (Figure 3). The r2 (goodness-of-fit) of the SHEI and kappa coefficient is 0.201, 0.316, 0.189 in 1996, 2006, 2016, respectively. It shows significant differences than others. There is an obvious correlation between kappa coefficient and SHEI with the trend being the same for different years. The correlation analysis results also prove that there is a significant correlation between kappa coefficient and SHEI (r = −0.562, p = 0.0151). Although the trends between NP and kappa coefficient of different years were same, the result is obviously affected by the outliers. Correlation analysis also indicates that there is no correlation among them. The relationship trends between other landscape indices and kappa coefficient of different years are inconsistent. Scatter plots of kappa coefficient and population and population density and per capita income were also marked (Figure 4). It can be found that population density has a more significant correlation with kappa coefficient than population and income (r = −0.269, p = 0.2812).
In recent years, many studies have used Markov and CA models to simulate land use. Gong et al. (2015) analyzed spatial and temporal changes of land use in Harbin from 1989 to 2007 and then to predict future trends by using CA-Markov model . Hyandye and Martz (2017) simulated Usangu Catchment’s LULC of 2020 based on LULC of 2000, 2006 and 2013 using Markov chain and CA analysis and analyzed the importance of social, edaphic, climatic and landscape geomorphology factors to LULC change . The simulation model of land use be affected by various factors. Spatial accessibility factors [53,54], topography, people density, housing density, land price, etc., have been investigated in previous records [22,45,55,56,57]. Especially, socio-economic data have a confirmatory impact to the simulation of the CA-Markov model . However, little information has been focused on landscape metrics. This study analyzed the change and simulation of land use in Anren County. The relationship between simulation accuracy and landscape metrics and socio-economic factors is studied. The timescale of this study is short and human factors play a major role, so the impact of landscape metrics and socio-economic factors on simulation accuracy was analyzed. By plotting the scatter plots, it was found that SHEI and population density have obvious correlation with the kappa coefficient of land use in the simulations of Anren County. Correlation analysis also confirmed that they were negatively correlated. Shannon’s evenness index is a landscape metric that describes the distribution uniformity of different landscape types and can reflect the landscape diversity of the study area. The more complex the land use in the study area, the more difficult it is to simulate land use based on the land use simulation principle, and the lower the accuracy. The population factor is one of the most important factors affecting land use and landscape pattern in human factors . The increase in population has led to raising in demand for food, housing, industrial output, etc. Causing changes in the structure and mode of land use further affects the diversity of land use. The greater the population density, the more complicated the land use. The population density is also a significant factor affecting simulation in many studies [22,55,56,59]. Therefore, both population density and kappa coefficient have a correlation that cannot be ignored. On the other hand, the weak correlation between kappa coefficient, landscape pattern and economic factors enhanced that the CA-Markov model is a powerful tool to predict the change of land use on county scale.
Over the past 30 years, land use types in Anren County have changed. The forest land and cultivated land were the dominant land types during this period. The area of forest land was slightly reduced. Changes in land use types have led to changes in landscape patterns, including patch and class levels. The CA-Markov model is a powerful tool to predict the change of land use on county scale. The relationship between simulation accuracy, landscape metrics and socio-economic factors is studied shows that SHEI and population density having an important impact on the accuracy of model predictions were negatively correlated with kappa coefficient. Through this study, we have found that it is difficult to understand how focused factors affect prediction accuracy of CA-Markov model, because the correlation analysis cannot reflect causality. In further studies, we should focus on the mechanism of external factors affecting the model.
This research was supported by Science and Technology Project of Yiyang City (2017JZ07), Key Technology R&D Program of Hunan Province (2016TP2007, 2017TP2006) and Major Science and Technology Program of Hunan Province (2017NK1014).
Behera MD, Borate SN, Panda SN, Behera PR, Roy PS. Modelling and analyzing the watershed dynamics using cellular automata (CA)-Markov model – A geo-information based approach. J Earth Syst Sci. 2012;121:1011–24. .
Yu W, Zang S, Wu C, Liu W, Na X. Analyzing and modeling land use land cover change (LUCC) in the Daqing City, China. Appl Geogr. 2011;31:600–8. .
Plieninger T, Draux H, Fagerholm N, Bieling C, Bürgi M, Kizos T, et al. The driving forces of landscape change in Europe: A systematic review of the evidence. Land Use Pol. 2016;57:204–14. .
Bürgi M, Hersperger AM, Schneeberger N. Driving forces of landscape change - current and new directions. Landsc Ecol. 2005;19:857–68. .
Lambin EF, Turner BL, Geist HJ, Agbola SB, Angelsen A, Bruce JW, et al. The causes of land-use and land-cover change: moving beyond the myths. Glob Env Change. 2001;11:261–9. .
Song X, Yang G, Yan C, Duan H, Liu G, Zhu Y. Driving forces behind land use and cover change in the Qinghai-Tibetan Plateau: a case study of the source region of the Yellow River, Qinghai Province, China. Env Earth Sci. 2009;59:793–801. .
Halmy MWA, Gessler PE, Hicke JA, Salem BB. Land use/land cover change detection and prediction in the north-western coastal desert of Egypt using Markov-CA. Appl Geogr. 2015;63:101–12. .
Velazquez A, Duran E, Ramirez I, Mas JF, Bocco G, Ramirez G, et al. Land use-cover change processes in highly biodiverse areas: the case of Oaxaca, Mexico. Glob Env Change. 2003;13:175–84. .
Zhang Z, Zinda JA, Yang Z, Yin M, Yu Q. Effects of topographic attributes on landscape pattern metrics based on redundancy ordination gradient analysis. Landsc Ecol Eng. 2018;14:67–77. .
Abdolalizadeh Z, Ebrahimi A, Mostafazadeh R. Landscape pattern change in Marakan protected area, Iran. Reg Viron Change. 2019;19:1683–99. .
Wang B, Li Y, Wang S, Liu C, Liu L. Oasis landscape pattern dynamics in Manas River Watershed based on remote sensing and spatial metrics. J Indian Soc Remote Sens. 2018;47:153–63. .
Li Z, Xie C, He X, Guo H, Wang L. Dynamic changes of landscape pattern and vulnerability analysis in Qingyi River Basin. EES. 2017;94:012189. .
Dai E, Wu Z, Du X. A gradient analysis on urban sprawl and urban landscape pattern between 1985 and 2000 in the Pearl River Delta, China. Front Earth Sci. 2018;12:791–807. .
Schulz C, Koch R, Cierjacks A, Kleinschmit B. Land change and loss of landscape diversity at the Caatinga phytogeographical domain – Analysis of pattern-process relationships with MODIS land cover products (2001–2012). J Arid Env. 2017;136:54–74. .
- Export Citation
Schulz C, Koch R, Cierjacks A, Kleinschmit B. Land change and loss of landscape diversity at the Caatinga phytogeographical domain – Analysis of pattern-process relationships with MODIS land cover products (2001–2012). J Arid Env. 2017;136:54–74.)| false 10.1016/j.jaridenv.2016.10.004.
Cabral AIR, Costa FL. Land cover changes and landscape pattern dynamics in Senegal and Guinea Bissau borderland. Appl Geogr. 2017;82:115–28. .
Hyandye C, Martz LW. A Markovian and cellular automata land-use change predictive model of the Usangu Catchment. Int J Remote Sens. 2016;38:64–81. .
Mei Z, Hao WU, Shiyun LI. Simulating land-use changes by incorporating spatial autocorrelation and self-organization in CLUE-S modeling: a case study in Zengcheng District, Guangzhou, China. Front Earth Sci. 2018;12:299–310. .
Wang Q, Liu R, Cong M, Guo L. Application of genetic algorithm to land use optimization for non-point source pollution control based on CLUE-S and SWAT. J Hydrol. 2018;560:86–96. .
Kundu S, Khare D, Mondal A. Individual and combined impacts of future climate and land use changes on the water balance. Ecol Eng. 2017;105:42–57. .
Ralha CG, Abreu CG, Coelho CGC, Zaghetto A, Macchiavello B, Machado RB. A multi-agent model system for land-use change simulation. Env Model Softw. 2013;42:30–46. .
Yang J, Liu W, Li Y, Li X, Ge Q. Simulating Intraurban land use dynamics under multiple scenarios based on fuzzy cellular automata: A case study of Jinzhou District, Dalian. Complexity. 2018;2018:1–17. .
Antoni JP, Vuidel G, Omrani H, Klein O. Geographic cellular automata for realistic urban form simulations: how far should the constraint be contained? In: D'Acci L, editor. The mathematics of urban morphology. Cham: Springer International Publishing; 2019. p. 147–62.
- Export Citation
Antoni JP, Vuidel G, Omrani H, Klein O. Geographic cellular automata for realistic urban form simulations: how far should the constraint be contained? In: D'Acci L, editor. The mathematics of urban morphology. Cham: Springer International Publishing; 2019. p. 147–62.)| false 10.1007/978-3-030-12381-9_7
Mukherjee AB, Krishna AP, Patel N. Application of remote sensing technology, GIS and AHP-TOPSIS model to quantify urban landscape vulnerability to land use transformation. In: Mishra D, Nayak M, Joshi A, editors. Information and communication technology for sustainable development. Singapore: Springer Singapore; 2018. p. 31–40.
Worku T, Khare D, Tripathi SK. Modeling runoff–sediment response to land use/land cover changes using integrated GIS and SWAT model in the Beressa watershed. Env Earth Sci. 2017;76:550. .
Klosterman RE, Koomen E, Stillwell J, Bakema A, Scholten HenkJ. Modelling land-use change: progress and applications. Appl Spatial Anal. 2008;1:151–2. .
Yang X, Zheng XQ, Lv LN. A spatiotemporal model of land use change based on ant colony optimization, Markov chain and cellular automata. Ecol Model. 2012;233:11–9. .
Sakieh Y, Amiri BJ, Danekar A, Feghhi J, Dezhkam S. Simulating urban expansion and scenario prediction using a cellular automata urban growth model, SLEUTH, through a case study of Karaj City, Iran. J Hous Built Env. 2014;30:591–611. .
Wang SQ, Zheng XQ, Zang XB. Accuracy assessments of land use change simulation based on Markov-cellular automata model. Procedia Env Sci. 2012;13:1238–45. .
Sang L, Zhang C, Yang J, Zhu D, Yun W. Simulation of land use spatial pattern of towns and villages based on CA–Markov model. Math Comput Model. 2011;54:938–43. .
White R, Engelen G. Cellular automata and fractal urban form: A cellular modelling approach to the evolution of urban land-use patterns. Env Plann A. 1993;25:1175–99. .
Liu X, Li X, Liu L, He J, Ai B. A bottom‐up approach to discover transition rules of cellular automata using ant intelligence. Int J Geogr Inf Sci. 2008;22:1247–69. .
Li X, Yeh AGO. Urban simulation using principal components analysis and cellular automata for land-use planning. Photogramm Eng Remote Sens. 2002;68:341–52. .
Takeyama M, Couclelis H. Map dynamics: integrating cellular automata and GIS through Geo-Algebra. Int J Geogr Inf Syst. 1997;11:73–91. .
Santé I, García AM, Miranda D, Crecente R. Cellular automata models for the simulation of real-world urban processes: A review and analysis. Landsc Urban Plan. 2010;96:108–22. .
Houet T, Hubert-Moy L. Modeling and projecting land-use and land-cover changes with Cellular Automaton in considering landscape trajectories. EARSeL eProceedings. 2006;5:63–76.
Fan F, Wang Y, Wang Z. Temporal and spatial change detecting (1998–2003) and predicting of land use and land cover in core corridor of Pearl River Delta (China) by using TM and ETM + images. Env Monit Assess. 2008;137:127–47. .
Jafari M, Majedi H, Monavari SM, Alesheikh AA, Zarkesh MK. Dynamic simulation of urban expansion through a CA-Markov model Case study: Hyrcanian region, Gilan, Iran. Eur J Remote Sens. 2017;49:513–29. .
López E, Bocco G, Mendoza M, Duhau E. Predicting land-cover and land-use change in the urban fringe: A case in Morelia city, Mexico. Landsc Urban Plan. 2001;55:271–85. .
He C, Okada N, Zhang Q, Shi P, Li J. Modelling dynamic urban expansion processes incorporating a potential model with cellular automata. Landsc Urban Plan. 2008;86:79–91. .
Liu X, Li X, Yeh AGO, He J, Tao J. Discovery of transition rules for geographical cellular automata by using ant colony optimization. Sci China, Ser D: Earth Sci. 2007;50:1578–88. .
Aitkenhead MJ, Aalders IH. Predicting land cover using GIS, Bayesian and evolutionary algorithm methods. J Env Manage. 2009;90:236–50. .
Oliveira Barros K, Alvares Soares Ribeiro CA, Marcatti GE, Lorenzon AS, Martins de Castro NL, Domingues GF, et al. Markov chains and cellular automata to predict environments subject to desertification. J Env Manage. 2018;225:160–7. .
Gidey E, Dikinya O, Sebego R, Segosebe E, Zenebe A. Cellular automata and Markov Chain (CA_Markov) model-based predictions of future land use and land cover scenarios (2015–2033) in Raya, northern Ethiopia. Model. Earth Syst Env. 2017;3:1245–62. .
- Export Citation
Gidey E, Dikinya O, Sebego R, Segosebe E, Zenebe A. Cellular automata and Markov Chain (CA_Markov) model-based predictions of future land use and land cover scenarios (2015–2033) in Raya, northern Ethiopia. Model. Earth Syst Env. 2017;3:1245–62.)| false 10.1007/s40808-017-0397-6.
Limited I. S., Image Processing Course for ENVI. 2014
Chukwudi N, Vilém P, Vit V. Responses of soil and plants to spatio-temporal changes in landscape under different land use in Imo watershed, southern Nigeria. Arch Agron Soil Sci. 2019;65:1–17. .
Zhang H-b, Zhao S-x. Development and application of land-use planning management information system based on ArcGIS. Proceedings of the 2010 International Forum on Information Technology and Applications - Volume 03. IEEE Computer Society; 2010. p. 64–7.
Mcgarigal KS, Cushman SA, Neel MC, Ene E, FRAGSTATS: Spatial pattern analysis program for categorical maps; 2002
Altuwaijri HA, Alotaibi MH, Almudlaj AM, Almalki FM. Predicting urban growth of Arriyadh city, capital of the Kingdom of Saudi Arabia, using Markov cellular automata in TerrSet geospatial system. Arab J Geosci. 2019;12:135. .
Gong W, Yuan L, Fan W, Stott P. Analysis and simulation of land use spatial pattern in Harbin prefecture based on trajectories and cellular automata—Markov modelling. Int J Appl Earth Obs. 2015;34:207–16. .
Feng Y, Yang Q, Hong Z, Cui L. Modelling coastal land use change by incorporating spatial autocorrelation into cellular automata models. Geocarto Int. 2016;33:470–88. .
Yao F, Hao C, Zhang J. Simulating urban growth processes by integrating cellular automata model and artificial optimization in Binhai New Area of Tianjin, China. Geocarto Int. 2015;31:612–27. .
Hu X, Li X, Lu L. Modeling the land use change in an arid oasis constrained by water resources and environmental policy change using cellular automata models. Sustainability. 2018;10:2878. .
Losiri C, Nagai M, Ninsawat S, Shrestha R. Modeling urban expansion in bangkok metropolitan region using demographic–economic data through cellular automata-Markov Chain and multi-layer perceptron-Markov chain models. Sustainability. 2016;8:686. .
de Oliveira Barros K, Alvares Soares Ribeiro CA, Marcatti GE, et al. Markov chains and cellular automata to predict environments subject to desertification. J Env Manage. 2018;225:160–7. .
Mondal B, Das DN, Bhatta B. Integrating cellular automata and Markov techniques to generate urban development potential surface: a study on Kolkata agglomeration. Geocarto Int. 2016;32:401–19. .
Fu X, Wang X, Yang YJ. Deriving suitability factors for CA-Markov land use simulation model based on local historical data. J Env Manage. 2018;206:10–9. .