Prediction of Flood-Prone zones based on Cellular Automata in GIS

Prediction of Flood-Prone zones based on Cellular Automata in GIS | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Prediction of Flood-Prone zones based on Cellular Automata in GIS rouzbeh shad, Seyed Mojtaba Mousavi, Marjan Ghaemi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3851820/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Due to climate change and rapid urbanization, urban flooding is on the rise, necessitating effective flood control measures in urban areas. Predicting potential flood-prone areas undergoing Land Use (LU) changes could significantly aid in planning for risk reduction and sustainable urban design. However, there's a scarcity of studies that consider both climate change and LU alterations. This research introduces a novel basin-scale framework utilizing a Future LU Simulation (FLUS) model to evaluate disaster-prone areas' risk from 20-year flood scenarios projected for 2040 and 2060. The Markov-FLUS model was developed and validated using historical data from 2000 to 2020. This model was then employed to simulate LU changes from 2020 to 2060 based on natural scenarios. Focusing on Khorasan Razavi as a case study, it investigates the potential consequences of LU transformations due to ongoing urbanization and vegetation changes in connection with predicted environmental shifts. The findings indicate an anticipated increase in accident-prone areas and constructed land in the studied area in the future. Spatially, this heightened flood risk primarily occurs on the periphery of existing developed areas or converted land. This framework's insights into future flood-prone areas' spatio-temporal characteristics offer valuable guidance for implementing rational flood mitigation measures in the most critical regions for development. Land use (LU) simulation Cellular automata (CA) Markov chains ANN FLUS model Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Repetitive natural occurrences like floods lead to notable economic losses and fatalities (Costache & Bui, 2019 ). Urban regions experiencing robust economic growth bear considerable effects from flood-related risks. Recent environmental shifts (climate change, land use change, etc.), have amplified the risk of floods in cities (Lin et al., 2022 ; Pour et al., 2020 ). Vulnerability refers to the inherent qualities of urban communities concerning hazards, covering aspects like people, infrastructure, and economic activities, influencing their susceptibility or capacity to manage these risks 'Exposure' describes (Parry et al., 2007 ). It closely relates to the concept of a flooding 'pathway' (DEFRA & EA, 2006), which considers both the location and specific features of urban regions that can either worsen or alleviate the impact of hazards. Risks emerge when urban communities and hazards align spatially, and the severity of these risks depends on the communities' vulnerability, the type of hazard, and the environmental characteristics involved (Lindley et al., 2006 ). However, forecasting flood risks is intricate, involving numerous factors like terrain, river dynamics, precipitation, and land-use (Shafizadeh-Moghadam et al., 2018 ). The Cellular Automata (CA) model serves as a spatial simulation tool essential for replicating land use (LU) patterns and configurations ( Gilpin, 2019 ). It plays a crucial role in studying the spatial alterations in LU, influenced by both natural phenomena and human factors ( Mor et al., 2021 ). Utilizes a future land-use simulation (FLUS) model to conduct computational evaluations of flood risk based on different scenarios. This FLUS model synergizes an artificial intelligence technique, specifically an artificial neural network (ANN), with a CA model for interactive operations. (Li et al., 2017 ). A pressing need exists to holistically assess various urbanization and dynamic flood risk scenarios concurrently. This research pairs the FLUS and floodplain inundation models how different urban development and climate change scenarios interact. The FLUS model, enhancing precision in simulations, integrates ANN and CA models to replicate nonlinear alterations in land use while factoring in environmental and societal elements (Liu et al., 2017 ; Zhai et al., 2020 ). In study (Yuqin et al., 2019), precipitation and LU indicators were analyzed using the (CA) method across the years 1986 to 2010 in the Qinhuai River basin. This study introduces a novel space-time framework to predict daily precipitation index at specific spatial extents. These models are integral components of stochastic weather generators, frequently employed in hydrological design or as inputs for environmental and ecosystem models (Richardson, 1981; Wilks & Wilby, 1999; Srikanthan & Wilby, 2001). Efficient examination and surveillance of land cover necessitate extensive data on Earth's surface and habitats. Often, human activities trigger changes in land use and land cover, which are subsequently influenced by natural processes. Conversely, the land use index stands out as a crucial variable determining resource planning and control strategies (Singh et al., 2015 ). Various elements regarding the Normalized Difference Vegetation Index (NDVI) and its associated methodologies and procedures have been discussed (Rouse et al., 1974 ). Developing robust methodologies for analyzing NDVI time series data remains a significant challenge in the remote sensing domain. In recent years, several models tailored for modeling NDVI time series have been introduced (Faith, 2015). This research creates a simulation technique for LU alterations in the Khorasan Razavi region from 2040 to 2060. Its goal is to construct a unified model chain that incorporates both LU shifts and vulnerability. The amalgamated FLUS chain, spatialization, and flood risk models are suggested to dynamically evaluate forthcoming flood risk and its correlation with societal progress. When tested in the Kashfroud basin, this structure exposes variations in flood risk due to LU changes between 2020 and 2060, providing guidance for future flood prevention strategies amid evolving social circumstances. 2. Data and Methods 2.1. Study Area Khorasan-Razavi province, situated in northeastern Iran, lies amidst two mountain ranges, Hazar Mosque to the north and Binaloud to the south. The region is crossed by the Kashfroud River, referred to as the Kashfroud catchment area. The study area is positioned between longitudes E58˚30˚-61˚ and latitudes N35˚30˚-37˚ within Khorasan province. The Kashfroud River plays a vital role as the primary water source for the Mashhad Plain, spanning approximately 250 km from its origin to its outlet and comprised of 25 branches. This area experiences predominantly hot and arid weather conditions, with lower altitudes compared to northern Khorasan. The vegetation primarily comprises plants requiring minimal water due to limited rainfall in the region (mousavi & arian, 2015 ; davari et al., 2018 ). Given the locations within the Kashfroud River vicinity, the region has seen continuous and seasonal activity throughout various cultural settlements. The total area spans approximately 8996 square kilometers, with 60% comprising mountainous terrain and 40% plains. The Mashhad plain, situated within relatively expansive valleys, extends up 28 to 30 km around Mashhad. Altitudes within the plain range from 890 to 1800 meters above sea level, with the highest point being the peak of the Binaloud mountain range, reaching 3287 meters, and the Hazar Mosque mountain range peak at 3150 meters. The area's lowest point, situated at the outlet, stands at 891 meters above sea level (davari et al., 2020 ; Azamirad & Esmaili, 2018 ). The basin's rivers experience seasonal fluctuations in response to the region's climate, occasionally leading to hazardous floods. Recent droughts and excessive human utilization of water resources beyond the basins' eco-hydrological capacity have created an adverse environmental condition within the Kashfroud watershed (Abedini & Pour Farrash Zadeh, 2021). 2.2. Data Resources and Processing Steps In assessing flood risk, the initial step involves building a spatial database encompassing various flood factors. However, the selection of suitable flood factors differs based on specific area characteristics, with these factors exhibiting distinct impacts across different regions. In this study, the Kashafrud Basin's features were considered, drawing from prior research to choose two data aspects: disaster-causing factors and disaster-prone environments. Landsat_8 OLI + TIRS/ETM + remote sensing image data from 2000 and 2020 were collected, while Landsat 7 and 8 remote sensing images were utilized for data interpretation in 2040 and 2060 (Xu et al., 2017 ). The analysis utilized LU data as its basis, which entailed the manual correction of raw LU data. Following data correction and manual interpretation, the classification scheme presented by Sciences was implemented, categorizing LU into six monitored classes: residential areas, non-residential areas, vegetation, rivers, mountains, and flood-prone areas (Liu et al., 2003 ). This data model can be accessed using the site https://eos.com/landviewer/?tool-timelapse . The mathematical challenges within diverse remote sensing data analysis techniques can be viewed as distinct instances of input-output data amalgamated with a specific network reliant on algorithms(Zhang et al., 2016 ). The Normalized Difference Vegetation Index (NDVI) functions as a standard gauge for generating images that showcase the verdant or relative biomass status of an area. This scale spans from − 1.0 to 1.0, predominantly reflecting vegetation. Negative values typically stem from clouds, water, or snow, while values near zero often denote rocks or barren soil. Leveraging the vegetation data within the Landsat imagery, efforts were made to minimize errors when applying the NDVI to Landsat images from 2000 and 2020, aiming to assess LU in urban or urban sprawl contexts (Alsharif et al., 2022 ). The primary cause of flood hazards within the surveyed region is primarily attributed to intense rainfall, where higher precipitation indicates an increased risk of potential flooding. Precipitation was selected as a key factor contributing to flood hazards, particularly with a rise in the frequency of heavy rainfall events. In subsequent research, annual precipitation data from 2000 to 2020 was obtained using the website https://earthengine.google.com/ , and the data was downloaded at a resolution of 1*1 km (Wu et al., 2024 ).The Advanced Thermal Reflectance Digital Earth Model (DEM) and Space Emission Radiometer (ASTER) with a 30-meter resolution are used as fundamental data sources for land elevations and slopes (Xu et al., 2017 ). The disaster-breeding environments incorporated seven elements: topographical aspects (DEM, slope, hillshade, aspect, flow accumulation, flow direction), along with Euclidean distance, among other spatial factors crucial in evaluating flood risk in Khorasan Razavi (Mohamoud, 1992 ; Vojtek & Vojteková, 2019 ). The digital elevation model (DEM) illustrates the elevations within the Kashfroud basin, inversely correlated with flood risk arising from water flow between different elevations. Obtained from https://search.earthdata.nasa.gov/search for this study, the DEM indicates topographic undulations, with slope significantly influencing flood occurrence (Costache & Bui, 2020 ). Slope directly impacts surface runoff and vertical percolation, affecting water accumulation, particularly in areas with low slopes that pose higher flood risks. ArcGIS was employed to calculate slope from the DEM dataset. The multiple flow direction algorithm allocates the flow to neighboring downslope cells based on slope, following a weighted distribution (Quinn et al., 1991 , 1995 ).Flow accumulation plays a crucial role in grasping the topographic influence on water. The DEM has been instrumental in automatically extracting drainage networks. Employing a fully parallel algorithm, we calculate the comprehensive flow directions ( Do et al., 2011 ). The utilization of DEM-based Hillshade enhances the depiction of elevation and topography within the area, offering valuable insights for land slope assessments, identification of peaks and valleys, understanding hydrological systems, and other topographic characteristics. We affirm that producing and precisely interpreting shaded relief images can significantly aid in addressing diverse issues.( Bajabaa et al., 2014 ). Topographic elements such as aspect, obtained from field surveys or digital elevation models (DEM), are commonly applied in modeling species distributions. However, extending these models to predict across larger regions or in future scenarios encounters challenges due to the intricate interplay between slope, aspect, and their specific relationships ( Bennie et al., 2008 ). This approach typically generates spatial concentrations resembling distinct lines with increased realism. During floods, riverbanks become inundated, making the distance to the river (DR) crucial for assessing flood risk in specific areas. Calculated using Euclidean distance (Costache et al., 2017), DR is essential for evaluating inundation risks along riverbanks during flooding events.The impact factor data underwent normalization, scaling all values between 0 and 1 for subsequent data mining processes (Muis et al., 2016 ). 2.3. Future LU Simulation (FLUS) Model The FLUS model, a novel simulation framework for LU changes, comprises two primary components: (1) an ANN-based module estimating the probability of occurrence and (2) a (CA) module incorporating a self-adaptive inertial competition mechanism using roulette selection. This model integrates factors like probability of occurrence, neighborhood influence, conversion cost, and inertia coefficient to compute the total conversion probability for each LU type within cells (Liu et al., 2017 ).Following this, the roulette mechanism is employed to portray competition among diverse LU types, determining their conversion and distribution on cells for a more accurate simulation of LU changes. This process is typically estimated using Eq. (1) (Liang et al., 2018 ; Liu et al., 2017 ): \({TP}_{t(m,n)}={P}_{m,n}\times {\varOmega }_{t(m,n)}\times Inertiat n\times \left(1-scc\to n\right)\) ( 1 ) \({\text{T}\text{P}}_{\text{t}(\text{m},\text{n})}\) refers to the likelihood of grid cell m transitioning from its initial landscape type to the target landscape type n at iteration time t ; \({\text{P}}_{\text{m},\text{n}}\) represents the probability of occurrence of landscape type n on grid cell m; \({{\Omega }}_{\text{t}(\text{m},\text{n})}\) signifies the impact of landscape type n's neighborhood on grid cell m during iteration time t; Inertiat n denotes the inertia coefficient of landscape type n at iteration time t, and scc→n indicates the conversion cost from the initial landscape type c to the target landscape type n. 2.3.1 ANN The simulation primarily assesses the likelihood of transforming various land-use categories. This process employs a neural network model comprising three layers: input, hidden, and output layers. By randomly sampling data from diverse spatial driving factors and multiple-factor urban land-use distribution layers, the artificial neural network (ANN) is trained to determine the probability of converting different land-use types, effectively estimating the suitability probability for expanding cell use. The operational functionality of the network as depicted in Fig. 2 . ( Cai et al., 2023 ). The initial module selected 11 influential factors to establish the Artificial Neural Network (ANN) model for the probability of occurrence concerning each LU type. The chosen factors encompassed both natural aspects like DEM, slope, hillshade, flow accumulation, NDVI, flow direction, and aspect, along with precipitation and distance metrics such as distance to rivers or Euclidean Distance. ANN, a model inspired by biological neural network simulations, handles complex nonlinear functions with numerous variables. Through iterative learning and recall processes, it fits intricate relationships between input data and training objectives, ensuring the generation of more accurate suitability probability distributions. This process establishes the correlation between the probability of each land type and the associated driving factors, expressed in Eq. (2) as: The suitability probability calculation is represented in the equation as: \({p}_{m,n }=p\left(m,n,t\right)= \sum _{i}{w}_{i,n }\times sigmoid\left({net}_{i}\left(m,t\right)\right)= \sum _{i}{w}_{i,n} \times \frac{1}{1+ {e}^{{net}_{i}\left(m,t\right)}}\) ( 2 ) This equation defines p(m, n, t) as the suitability probability of landscape type non-grid cell m at time t. Here, \({\text{w}}_{\text{i},\text{n} }\) and sigmoid denote the weight and activation function between the hidden layer and the output layer, respectively. Additionally, \({\text{n}\text{e}\text{t}}_{\text{i}}\left(\text{m},\text{t}\right)\) represents the signal received from grid m at time t in the hidden layer (Liu et al., 2017 ). 2.3.2 CA–Markov Model The CA model functions as a spatial dynamic model with discrete attributes in time, space, and states, integrating local spatial interactions and temporal cause-and-effect relationships ( Wolfram, 1983) ,Operating on a structured grid system, each cell possesses a finite set of discrete states and neighboring cells, where its future state is determined by its own state and the states of neighboring cells in the previous time step. Employing rules governing cell state changes enables the simulation of spatial and temporal changes in LU and various intricate systems (Li & yeh, 2002).The model representation is depicted as follows ( Wolfram, 1983): \(S\left(t+1\right)=F\left(S\left(t\right),N\right)\) ( 3 ) S(t) and S(t + 1) represent the collection of cell states at times t and t + 1 correspondingly; F denotes the regulation governing cellular state transitions, while N signifies the neighborhood filter. The Markov model stands as a classical statistical and quantitative forecasting technique ( Wijesekara et al., 2012 ), Developed by Andrey Markov from the former Soviet Union, this method finds application across various fields like natural language processing, human resource management, and LU simulation, among others. \(S\left(t+1\right)=P\times S\left(t\right)\) ( 4 ) \(P=\left[\begin{array}{c}{p}_{11} {p}_{12} \dots {p}_{1n}\\ {p}_{21} {p}_{22} \dots {p}_{2n}\\ . . . .\\ . . . .\\ . . . .\\ {p}_{n1} {p}_{n2} \dots {p}_{nn}\end{array}\right]\) ( 5 ) where S(t) and S(t + 1) are the LU states at times t and t + 1, respectively; P is the state transition probability matrix; \({p}_{ij}\) is the probability of class i land transforming into $$\text{c}\text{l}\text{a}\text{s}\text{s} \text{j} \text{l}\text{a}\text{n}\text{d},{\text{p}}_{\text{i}\text{j}}\in \left(\text{0,1}\right),\text{a}\text{n}\text{d}\sum _{\text{j}=1}^{\text{n}}{\text{p}}_{\text{i}\text{j}}=1,\left(\text{i},\text{j}=\text{1,2},3\dots ,\text{n}\right). \left(6\right)$$ In LU simulation, the Markov model concentrates on predicting change quantities but lacks the ability to portray the spatial distribution or patterns of various land changes ( Nor et al., 2017 ) Conversely, the CA model forecasts the spatial-temporal characteristics of diverse land changes, addressing the Markov model's limitations ( Wijesekara et al., 2012 ). The model is depicted as follows: $${\text{s}}_{\text{i}\text{j}}\left(\text{t}+1\right)=\text{F}({\text{s}}_{\text{i}\text{j}}\left(\text{t}\right),{\text{Q}}_{\text{i}\text{j}}\left(\text{t}\right),\text{V})$$ 7 \({s}_{ij}\) (t) and \({s}_{ij}\) (t + 1) represent the conditions of the cell in row i and column j at times t and t + 1, respectively; \({Q}_{ij}\) (t) denotes the state of the neighboring cells of the cell in row i and column j at time t; V represents the suitability atlas; and F is the rule governing cell transformations. 2.3.3 Kappa Coefficient The Kappa coefficient is commonly employed to assess the precision of the anticipated LU map. Typically, a Kappa value surpassing 0.75 is considered indicative of superior accuracy in the prediction outcome ( Araya & cabral, 2010 ; Tong et al., 2012 ). If the total number of pixels is N, and S represents the count of accurately simulated grids, where the actual counts for each class are a1, a2,. . ., ac, and the simulated counts for each class are b1, b2,. . ., bc, respectively, the Kappa coefficient (K) can be mathematically represented using the following formulas: $$\text{K}=\frac{{\text{P}}_{0}-{\text{P}}_{\text{c}}}{1-{\text{P}}_{\text{c}}} \left(8\right)$$ $${\text{P}}_{0}=\frac{\text{S}}{\text{N}} \left(9\right)$$ $${\text{P}}_{\text{c}}=\frac{{\text{a}}_{1}\times {\text{b}}_{1}\times {\text{a}}_{2}\times {\text{b}}_{2}+\dots +{\text{a}}_{\text{c}}\times {\text{b}}_{\text{c}}}{\text{N}\times \text{N}} \left(10\right)$$ The general accuracy measures the proportion of correct predictions made by the model across all test sets against the total count. The model exhibits high accuracy when Kappa is ≤ 0.75 and an overall accuracy (OA) is less than 1, moderate accuracy when Kappa is ≤ 0.5 and OA is less than 0.75, and poor accuracy when Kappa is ≤ 0 and OA is less than 0.5. urthermore, the validation of the alterations was confirmed through the employment of the figure of merit (Fom), considered more effective than the Kappa coefficient in evaluating the precision of simulated alterations (Pontius et al. 2008). $$\text{F}\text{o}\text{m}=\frac{\text{B}}{\text{A}+\text{B}+\text{C}+\text{D}}\times 100\text{%} \left(11\right)$$ Here, A denotes an error area resulting from observed change predicted as persistence, B signifies an accurate area resulting from observed change predicted as change, C represents an error area resulting from observed change predicted as shifting to an incorrect category, and D indicates an error area resulting from observed persistence predicted as change. 3. Results and Analysis 3.1 Emporal Alterations in Land Utilization The forecast of future LU demand was conducted using the Markov chain technique.The pixel-based data has a resolution of 30 m. ArcGIS 10.5 facilitated the reclassification and graphical representation of LU maps in Khorasan Razavi for 2000 and 2020, The Markov chain module was employed to compute the LU requirements for different types in the years 2040 and 2060. (Fig. 4 ). To assess the demand for various LU types in 2040 and 2060, Lu data from 2000 and 2020 were incorporated into the Markov chain module. Details regarding the area of each LU and its evolving trend are outlined in (Table 1 ).The land types were mainly classified into established, comprising residential areas, non-residential areas, vegetation, rivers, mountains, and flood-prone areas. The area changes of the six types of LU for the towenty years were also counted. Table 1 LU classes distribution in area ( \({\text{k}\text{m}}^{2}\) ) in khorasan razavi from 2000 and 2020. Year Residential area Non residential area Vegetation Mountain River Accidental spots \({km}^{2}\) % \({km}^{2}\) % \({km}^{2}\) % \({km}^{2}\) % \({km}^{2}\) % \({km}^{2}\) % 2000 288.99 15.6 489.65 26.4 190.81 10.3 795.83 42.9 83.21 4.4 2.54 0.13 2020 316.35 17 469.68 25.3 152.46 8.2 818.62 44.2 91.01 4.9 3.05 0.16 LU data from 2000 and 2020 were used to determine the required parameters of the FLUS model. A significant expansion in residential areas, vegetation, accident-prone spots and Razavi rivers of Khorasan was observed when analyzing changes in latitude and time (Fig. 3 and Table 2 ). On the contrary, during this period, a decrease in the area covered by mountainous and non-residential areas was observed. These changes between 2000 and 2020 significantly affected the displacement and strengthening of vulnerable areas. To train the ANN model, 2056890 samples were randomly selected in the study area. To evaluate the accuracy of the FLUS model, it was used with a random sample, which resulted in a figure of merit (FOM) of 0.064 and a kappa coefficient of 0.94 for simulation. Despite the relatively low FoM, previous LU modeling experiments typically vary the reported values in the range of 0.1–0.3. This inconsistency is caused by the effects of path dependence, which causes inaccuracy in the prediction of land change models. Based on LU data from 2000 and 2020, the demand for different types of LU in 2040 and 2060 was predicted using the Markov chain modul. 3.2 Drivers of Land-Use Change The selection of driving factors hinges on research aims, regional traits, and data accessibility. The generation of suitability maps and the LU conversion probability matrix, as well as the transition suitability maps, heavily relies on these factors and limited areas. Figure 4 showcases spatial analyses concerning natural factors influencing urban land expansion. For this study's context, natural driving factors comprised DEM, slope, slope direction, aspect, hillshade, and annual precipitation. Additionally,Euclidean distance analysis is for rivers, flow accumulation, flow direction locations, and NDVI, regarded as ecological factors impacting LU transformation. These constituted the primary drivers of LU changes within this research. Spatial data derived from diverse sources, including geocoded location coordinates, underwent manipulations such as clipping and merging to align with the coverage area. Nine natural factors, delineated in Fig. 4 , served as the driving forces of change. These factors were defined based on the researchers' expertise in the study area, with weights derived from a series of paired comparisons. 3.3 Conversion Expenses and Limitations on Area Changes Conversion cost is utilized to quantify the level of complexity associated with transitioning from the current LU type to the intended type, playing a significant role in molding LU dynamics. Various factors influence the conversion processes between different LU types, typically including: (a) probability of occurrence, and (b) LU pair conversion cost, which signifies the challenge in shifting from one LU and Land Cover (LULC) class to another. This serves as a mathematical representation of the Earth's motion direction. In the cost matrix settings, a value of 0 denotes a higher land conversion cost and diminished potential for land expansion, while a value of 1 signifies a lower land conversion cost and higher potential for land expansion. Referring to the LU transition matrix from 2000 to 2020, the final cost matrix for LU change was established (Table 2 ). Table 2 Conversion cost coefficients between LU types. Residential area Non residential area vegetation mountain river Accidental spots Residential area 1 0 1 0 1 1 NON residential area 1 1 1 1 0 1 vegetation 0 0 1 1 1 1 mountain 1 1 1 1 1 1 river 0 1 1 1 1 Accidental spots 1 0 1 1 1 1 3.4 Neighborhood Factors Neighborhood factors indicate the capacity for different land types to expand, particularly their potential expansion under external influences. Ranging between 0 and 1, higher values signify a greater ability for a specific LU type to expand. These factors are determined by scrutinizing the LU data within the study area and integrating expert insights. Additionally, by scrutinizing the LU conversion matrix from 2000 to 2020 within the basin, the FLUS model's neighborhood weights were derived. Initial computations were based on the proportion of the extended area for each LU type, with continuous optimization and adjustment during the simulation process to obtain the optimal neighborhood weight parameters for the experiment, as depicted in Table 3 . Table 3 weight of Neighborhood parametrs. Land use types Residential area Non residential area Vegetation Mountain River Accidental spots Neighborhood weight 0.4 0.5 0.5 0.3 0.9 1 3.5 Simulation of LU Arrangement through FLUS Modeling This study utilized ArcGIS 10.5 to convert spatial data and standardized the data of Khorasan Razavi province, maintaining a resolution of 30 meters as the standard. Processing was conducted employing the GeoSOS-FLUS V2.4 model to simulate and forecast flood hotspots and their spatial fluctuations. 3.6 Calculation of Suitability Probability using Neural Network Techniques For this investigation, precipitation, NDVI, DEM, slope, direction, Euclidean distance, hill shadow, direct flow, and area flow accumulation were extracted as influential factors for floods and their locations. During the artificial neural network training with GeoSOS-FLUSV2.4, a uniform distribution sampling approach with a sampling ratio of 1% was adopted, configuring the hidden layer to consist of 13 nodes. Figure 6 displays the probability maps generated for each factor. The calculated root mean square error (RMSE) during training was 0.21, signifying a high level of training accuracy. The RMSE formula is as follows: $$\text{R}\text{M}\text{S}\text{E}=\sqrt{\frac{1}{\text{n}} \sum _{\text{i}=1}^{\text{n}}{\left(\widehat{{\text{y}}_{\text{i}}}-{\text{y}}_{\text{i}}\right)}^{2}} \left(12\right)$$ where n is the number of data points, \(\widehat{{\text{y}}_{\text{i}}}\) is the predicted value of the data point, and \({\text{y}}_{\text{i}}\) is the observed value of the data point. This study predicted the future changes in different types of LU by analyzing the classification data of the initial and final years and used the SD model for prediction. The relative influence of neighborhood factors was obtained from multiple tests, considering conversion rates among different LU types in the confusion matrix. After the finalization of the simulation, the output shows the results of the simulation. In addition, an estimation model using ANN was built based on the previously mentioned driving forces, as shown in Fig. 6 . This model effectively shows the probability of occurrence for each type of LU. 3.7 Validity of the Model Validation of the accuracy of FLUS model in predicting future exploitation in Razavi Khorasan province requires the verification of the previous model. Using LU data from 2000 as the initial dataset, the FLUS model simulated the 2020 LU pattern. The spatial image of simulated LU in 2020 in the study area is presented in (Fig. 7 ). The findings showed a strong alignment between actual and simulated LU patterns for 2020. Comparative analysis between real and simulated user maps enabled calculation of Kappa coefficient and FoM. The calculated values for the 2020 simulation results show a Kappa coefficient of 0.94 and a FoM of 0.064, indicating a satisfactory level of accuracy in the simulation. The accuracy assessment of the 2020 landscape pattern simulation included visual interpretations that showed a significant similarity between the predicted 2020 map and the actual map. Further visual analysis of the zoomed-in simulation results from three specific areas strengthened the FLUS model's ability to accurately predict future LU in the study area. 3.8 Change of Future Spatial-temporal Use in Khorasan Razavi By interpreting remote sensing images from Landsat 7 ETM + and Landsat-8 OLI TIRS and cross-verifying and rectifying the outcomes, the recent LU maps of the study zone for both 2000 and 2020, meeting the stipulated accuracy criteria (Fig. 7 ). In summary, vulnerable areas are projected to become larger, with an expected increase of 3.47 km2 by 2040 and 3.85 km2 by 2060, representing approximately 0.18% and 0.20% of the total area, respectively. Using the simulated model for LU change, based on the 2020 LU data as basic data and using the probability of each type of LU as automatic cell conversion rules, forecast maps for 2040 and 2060 were prepared. The maps show the areas under the natural development scenarios shown in Fig. 8 . These findings indicate a gradual expansion of flood-prone areas between 2040 and 2060 under the natural development scenario and we see more illegal constructions and agricultural lands in the river. Table 4 The amount of changes relative to the LU area in the study area in the future from 2020 to 2060 under the natural scenario. Year Residential area Non residential area Vegetation Mountain River Accidental spots \({km}^{2}\) % \({ km}^{2}\) % \({ km}^{2}\) % \({ km}^{2}\) % \({ km}^{2}\) % \({km}^{2}\) % 2020 S 316.35 17 469.68 25.3 152.44 8.2 818.50 44.2 91.00 4.9 3.05 0.16 2040 S 328.99 17.7 461.01 24.9 137.20 7.4 821.73 44.3 98.64 5.3 3.47 0.18 2060 S 334.78 18 456.54 24.6 130.99 7 818.69 44.2 106.2 5.7 3.85 0.20 4. Discussion Investigating the interaction between future urbanization trends and environmental transformations, including the increase of disaster prone areas due to climate change (rainfall) and changes in river boundaries, is very important to prevent unforeseen disasters. The relationship between these elements is complex and constantly changing. In our current review, the evaluation relied on a representative computing scenario. The method is robust and comprehensive, adept at handling the complexities of the factors involved, and proves to be an effective approach for determining the most accurate locations. Such identification serves as a basis for innovative planning and strategy aimed at reducing flood risk. 4.1. Limitations and Future works There are constraints within our research that necessitate refinement in future endeavors. Initially, there is ample room for enhancing the model's predictive precision. This study evaluates the performance of the FLUS model through Kappa and OA metrics, exhibiting a commendable level of predictive accuracy. Moreover, it establishes that 11 influential factors contribute to modeling and forecasting accident-prone areas in Khorasan Razavi. Integrating human and natural driving factors into LU change through the FLUS model, coupled with ANN, can yield more pragmatic simulation outcomes. However, the accuracy of predictions could be heightened by incorporating the latest land cover data. On one hand, this study is founded on DEM data modeling, which might result in underestimation or overestimation of simulation efficacy due to DEM data inaccuracies. On the other hand, amalgamating rainfall and land vegetation data bolsters the dependability of severe flood predictions. The conversion of other LUs into urban utilization substantially escalates flood hazards owing to alterations in impermeable surfaces. Consequently, in forthcoming forecasts, it is crucial to dynamically adjust pertinent factors influencing peak flows and flood risk to bolster forecast accuracy. Despite the inevitability of uncertainties in flood risk assessments, the divergence in outcomes produced by the proposed model remains within an acceptable range, ensuring the accuracy of flood risk evaluations. 4.2.Suggestions regarding Strategies and Policies for Urban Flooding In the 21st century, urban flooding poses a significant challenge for cities. Human responses to flood impacts heavily rely on how urban facilities are allocated and how managers plan for the future. The recent flooding in Khorasan Razavi province inflicted severe damage on both infrastructure and natural resources. These floods resulted in road destruction, loss of agricultural fields, and damage to residences and structures. The repercussions of these natural calamities resulted not just in financial setbacks but also in human, psychological, and societal losses. The enduring impact of these floods on people's lives and the rehabilitation of affected areas have presented numerous challenges for the local community. Hence, there's an urgent call for implementing flood risk adaptation strategies in Khorasan Razavi. 5. Conclusion This study introduces an approach using a real options strategy for urban planning and strengthening vulnerable sites against flood risks in the midst of environmental change. Various evaluations including overall accuracy, kappa coefficient and figure of merit (Fom) were performed in different programs. The simulated results for future LU show high simulation accuracy that aligns with research requirements. Considering the impact of urbanization and environmental changes, the model provides a significant insight into the extent and spatial expansion of flood hazards in Khorasan Razavi, which indicates a significant intensification of flooded urban areas in the future. The simulations predict an increase in the built-up area along the river banks in 2040 and 2060 compared to 2020. It is estimated that these constructions along the river will cover approximately 5.3 to 5.7% of the entire river due to the increase in water volume due to environmental changes. to increase by 3.85 square kilometers compared to 2020 and provide insights into flood risk areas. determined by simulation. This analysis deals with the spatial distribution of urban and rural areas prone to future floods and enables a comprehensive understanding of urban landscapes and challenges from the perspective of adaptive urban planning. It prioritizes flood mitigation strategies and highlights critical areas in the region. This paper introduces a new framework using the real options approach for planning and designing urban flood control measures and addresses the impact of environmental changes on urban flooding. However, there are areas for further improvement in this study. Considering the uncertainty in precipitation forecast due to global climate model accuracy, the use of dynamic downscaling methods can increase future precipitation forecasts. In addition, future research could explore additional measures for urban resilience to combat the effects of environmental change on urban flooding. Declarations Acknowledgments We extend our appreciation to the Water Management Institute of Iran for granting access to the flood database, facilitating data acquisition for this research. COI Statement On behalf of all authors, the corresponding author states that there is no conflict of interest. Declaration of Interests The authors have no relevant financial or non financial intest to disclose. Data Availibility Data and programs will be made available on request References Abedini M, Zadeh PF, F (2021) Analysis and Modeling of the Relationship between Monthly Discharge and Geomorphometric Characteristics (Case Study: Kashafrood Watershed). Geogr Environ Plann 32(4):29–44 Alsharif M, Alzandi AA, Shrahily R, Mobarak B (2022) Region Saudi Arabia Forests 13(10):1530Land Use Land Cover Change Analysis for Urban Growth Prediction Using Landsat Satellite Data and Markov Chain Model for Al Baha Araya YH, Cabral P (2010) Analysis and modeling of urban land cover change in Setúbal and Sesimbra, Portugal. Remote Sens 2(6):1549–1563 Arora A, Arabameri A, Pandey M, Siddiqui MA, Shukla UK, Bui DT, Bhardwaj A (2021) Optimization of state-of-the-art fuzzy-metaheuristic ANFIS-based machine learning models for flood susceptibility prediction mapping in the Middle Ganga Plain, India. Sci Total Environ 750:141565 Azamirad M, Esmaili K (2018) Investigation flooding potential in the Kashafrud watershed, Mashhad the method SCS and GIS. J Watershed Manage Res 9(17):26–38 Bajabaa S, Masoud M, Al-Amri N (2014) Flash flood hazard mapping based on quantitative hydrology, geomorphology and GIS techniques (case study of Wadi Al Lith, Saudi Arabia). Arab J Geosci 7:2469–2481 Bennie J, Huntley B, Wiltshire A, Hill MO, Baxter R (2008) Slope, aspect and climate: spatially explicit and implicit models of topographic microclimate in chalk grassland. Ecol Model 216(1):47–59 Cai Y, Zong W, Jiao S, Wang Z, Ou L (2023) Land-Use Assessment and Trend Simulation from a Resilient Urban Perspective: A Case Study of Changsha City. Sustainability 15(18):13890 Chapi K, Singh VP, Shirzadi A, Shahabi H, Bui DT, Pham BT, Khosravi K (2017) A novel hybrid artificial intelligence approach for flood susceptibility assessment, vol 95. Environmental modelling & software, pp 229–245 Costache R, Bui DT (2019) Spatial prediction of flood potential using new ensembles of bivariate statistics and artificial intelligence: A case study at the Putna river catchment of Romania. Sci Total Environ 691:1098–1118 Costache R, Bui DT (2020) Identification of areas prone to flash-flood phenomena using multiple-criteria decision-making, bivariate statistics, machine learning and their ensembles. Sci Total Environ 712:136492 Costache R, Pham QB, Avand M, Linh NTT, Vojtek M, Vojteková J, Dung TD (2020) Novel hybrid models between bivariate statistics, artificial neural networks and boosting algorithms for flood susceptibility assessment. J Environ Manage 265:110485 Davari MS, Basafa H, Toghraei M (2020) Kashafrud River (Mashhad plain) from the Neolithic to the Middle Bronze Age. J Archaeol Stud 12(3):41–63 Davari M, Mousivand F, Basafa H (2018) Explanation of Early Pottery Features in Kashafrud River Basin: Based on the typology of Pottery, s Esmail-Abad site in Mashhad. J Archaeol Stud 10(1):75–94 DEFRA, & EA (2006) R&D outputs: Flood risks to people. Phase 2. FD2321/TR1 The flood risks to people methodology. Department for Environment Food and Rural Affairs and the Environment Agency, London Do HT, Limet S, Melin E (2011) Parallel computing flow accumulation in large digital elevation models. Procedia Comput Sci 4:2277–2286 Gao Y, Chen J, Luo H, Wang H (2020) Prediction of hydrological responses to land use change. Sci Total Environ 708:134998 Gilpin W (2019) Cellular automata as convolutional neural networks. Phys Rev E 100(3):032402 Li X, Yeh AGO (2002) Neural-network-based cellular automata for simulating multiple land use changes using GIS. Int J Geogr Inf Sci 16(4):323–343 Li X, Chen G, Liu X, Liang X, Wang S, Chen Y, Xu X (2017) A new global land-use and land-cover change product at a 1-km resolution for 2010 to 2100 based on human–environment interactions. Annals of the American Association of Geographers 107(5):1040–1059 Liang X, Liu X, Li X, Chen Y, Tian H, Yao Y (2018) Delineating multi-scenario urban growth boundaries with a CA-based FLUS model and morphological method. Landsc urban Plann 177:47–63 Lin J, He P, Yang L, He X, Lu S, Liu D (2022) Predicting future urban waterlogging-prone areas by coupling the maximum entropy and FLUS model. Sustainable Cities and Society 80:103812 Lindley SJ, Handley JF, Theuray N, Peet E, McEvoy D (2006) Adaptation strategies for climate change in the urban environment: assessing climate change related risk in UK urban areas. J Risk Res 9(5):543–568 Liu J, Liu M, Zhuang D, Zhang Z, Deng X (2003) Study on spatial pattern of land-use change in China during 1995–2000. Sci China Ser D: Earth Sci 46:373–384 Liu J, Wang J, Xiong J, Cheng W, Cui X, He W, Wang N (2021) Dynamic Assessment of the flood risk at basin scale under simulation of land-use scenarios and spatialization technology of factor. Water 13(22):3239 Liu X, Liang X, Li X, Xu X, Ou J, Chen Y, Pei F (2017) A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects. Landsc Urban Plann 168:94–116 Mohamoud YM (1992) Evaluating Manning's roughness coefficients for tilled soils. J Hydrol 135(1–4):143–156 Mor B, Garhwal S, Kumar A (2021) A systematic review of hidden Markov models and their applications. Arch Comput Methods Eng 28:1429–1448 Mosavi E, Arian M (2015) Neotectonics of Kashaf Rud River, NE Iran by Modified Index of Active Tectonics (MIAT). Int J Geosci 6(07):776 Muis S, Verlaan M, Winsemius HC, Aerts JC, Ward PJ (2016) A global reanalysis of storm surges and extreme sea levels. Nat Commun 7(1):11969 Muriithi FK (2015) Centered log-ratio (clr) transformation and robust principal component analysis of long-term NDVI data reveal vegetation activity linked to climate processes. Climate 3(1):135–149 Nor ANM, Corstanje R, Harris JA, Brewer T (2017) Impact of rapid urban expansion on green space structure. Ecol Ind 81:274–284 Parry ML, Canziani OF, Palutikof JP, Linden PJ, Hanson CE (2007) IPCC Fourth Assessment Report: Climate Change 2007. The Fourth Assessment Report of the Intergovernmental Panel on Climate Change Pour SH, Abd Wahab AK, Shahid S, Asaduzzaman M, Dewan A (2020) Low impact development techniques to mitigate the impacts of climate-change-induced urban floods: Current trends, issues and challenges. Sustainable Cities and Society 62:102373 Quinn PFBJ, Beven K, Chevallier P, Planchon O (1991) The prediction of hillslope flow paths for distributed hydrological modelling using digital terrain models. Hydrol Process 5(1):59–79 Quinn PF, Beven KJ, Lamb R (1995) The in (a/tan/β) index: How to calculate it and how to use it within the topmodel framework. Hydrol Process 9(2):161–182 Richardson CW (1981) Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resour Res 17(1):182–190 Rouse JW, Haas RH, Schell JA, Deering DW (1974) Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec Publ 351(1):309 Shafizadeh-Moghadam H, Valavi R, Shahabi H, Chapi K, Shirzadi A (2018) Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping. J Environ Manage 217:1–11 Singh SK, Mustak S, Srivastava PK, Szabó S, Islam T (2015) Predicting spatial and decadal LULC changes through cellular automata Markov chain models using earth observation datasets and geo-information. Environ Processes 2:61–78 Srikanthan R, McMahon TA (2001) Stochastic generation of annual, monthly and daily climate data: A review. Hydrol Earth Syst Sci 5(4):653–670 Tong ST, Sun Y, Ranatunga T, He J, Yang YJ (2012) Predicting plausible impacts of sets of climate and land use change scenarios on water resources. Appl Geogr 32(2):477–489 Vojtek M, Vojteková J (2019) Flood susceptibility mapping on a national scale in Slovakia using the analytical hierarchy process. Water 11(2):364 Wijesekara GN, Gupta A, Valeo C, Hasbani JG, Qiao Y, Delaney P, Marceau DJ (2012) Assessing the impact of future land-use changes on hydrological processes in the Elbow River watershed in southern Alberta. Can J hydrology 412:220–232 Wilks DS, Wilby RL (1999) The weather generation game: a review of stochastic weather models. Prog Phys Geogr 23(3):329–357 Wolfram S (1983) Statistical mechanics of cellular automata. Rev Mod Phys 55(3):601 Wu X, Shen X, Li J, Xie X (2024) Determination and projection of flood risk based on multi-criteria decision analysis (MCDA) combining with CA-Markov model in Zhejiang Province, China, vol 53. Urban Climate, p 101769 Xu XL, Liu JY, Zhang ZX, Zhou WC, Zhang SW, Li RD, Shi XZ (2017) A Time Series Land Ecosystem Classification Dataset of China in Five-Year Increments (1990. Journal of Global Change Data & Discovery2017, 1(1), 52–59 Zhai Y, Yao Y, Guan Q, Liang X, Li X, Pan Y, Zhou J (2020) Simulating urban land use change by integrating a convolutional neural network with vector-based cellular automata. Int J Geogr Inf Sci 34(7):1475–1499 Zhang L, Zhang L, Du B (2016) Deep learning for remote sensing data: A technical tutorial on the state of the art. IEEE Geoscience and remote sensing magazine 4(2):22–40 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3851820","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":268296025,"identity":"f823a93d-3c47-4d9e-bfb3-0d07762b8df8","order_by":0,"name":"rouzbeh shad","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApElEQVRIiWNgGAWjYHACNiCygbINiNeSRrqWwyS4yryB/dmDH2XnE9c2MD/8wFBwj7AWmQM85oY9524nbjvAZizBYFBMWIsEAw+bBG8bSAuDGdAvCcRoYX8m+bftHFAL+zditTCYSfO2HQBq4SHWFmYeM2mZc8nG2w7zFEskEKWFvf2Z5JsyO9ltx9s3fvjwhwgtDMzIDGI0jIJRMApGwSggAgAA5Xsv4bfjSe4AAAAASUVORK5CYII=","orcid":"","institution":"Ferdowsi University of Mashhad","correspondingAuthor":true,"prefix":"","firstName":"rouzbeh","middleName":"","lastName":"shad","suffix":""},{"id":268296026,"identity":"c89d7875-2605-41e5-b611-8068e386ae7b","order_by":1,"name":"Seyed Mojtaba Mousavi","email":"","orcid":"","institution":"Ferdowsi University of Mashhad Faculty of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Seyed","middleName":"Mojtaba","lastName":"Mousavi","suffix":""},{"id":268296027,"identity":"7c760d80-36bd-4684-87bd-ce1ff6fc3b91","order_by":2,"name":"Marjan Ghaemi","email":"","orcid":"","institution":"Ferdowsi University of Mashhad Faculty of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Marjan","middleName":"","lastName":"Ghaemi","suffix":""}],"badges":[],"createdAt":"2024-01-11 00:33:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3851820/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3851820/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50046971,"identity":"6a41c209-32f6-4a36-8f74-606dc9e1e45a","added_by":"auto","created_at":"2024-01-23 16:06:30","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":453844,"visible":true,"origin":"","legend":"\u003cp\u003eLocation map of the study area. The main internal rivers of Khorasan-Razavi pass through Mashhad city through the Kashaf-Rud River and after leaving, they join the Hariroud River.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/feef901b184b7d1ece641921.jpeg"},{"id":50045736,"identity":"4e5c38b2-219d-4a9b-9d9c-91d3f6489a9b","added_by":"auto","created_at":"2024-01-23 15:58:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":213381,"visible":true,"origin":"","legend":"\u003cp\u003eProbability-of-occurrence model.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/790bc563df0530b899a6d530.png"},{"id":50046970,"identity":"c8d6f10d-a4cd-4439-9e0e-fd760b219d40","added_by":"auto","created_at":"2024-01-23 16:06:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":26507,"visible":true,"origin":"","legend":"\u003cp\u003eLU change trend in khorasan razavi for 2000 and 2020.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/5ce41db7ad2b295117cde734.png"},{"id":50045731,"identity":"ddac8ff1-307f-484d-95e2-f4fdcfceb92e","added_by":"auto","created_at":"2024-01-23 15:58:30","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":630635,"visible":true,"origin":"","legend":"\u003cp\u003espatial and temporal changes of the landscape pattern of the studied area in the years (A) 2000 to (B) 2020.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/1891adac96b76bb120eaf9f7.jpeg"},{"id":50045733,"identity":"c86c872f-2443-48b9-8e89-283c0b40612f","added_by":"auto","created_at":"2024-01-23 15:58:30","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":577405,"visible":true,"origin":"","legend":"\u003cp\u003eDrivers of land-use change. (A) DEM, (B) Slope, (C) Aspect, (D) NDVI 2000, (E) NDVI 2020, (F) Flow direction, (G) Flow accumulation, (H) Precipitation 2000, (I) Precipitation 2020, (J) Flow accumulation, (K) Euclidean distance.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/b217a1412db0e8eeee28a8d3.jpeg"},{"id":50045735,"identity":"1e60f718-6bc5-4ee8-9242-da3dc50c3ad2","added_by":"auto","created_at":"2024-01-23 15:58:30","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":102768,"visible":true,"origin":"","legend":"\u003cp\u003eAAN-based suitability probability map of occurrence.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/8a51e7826a89c80fbbcf47e9.jpeg"},{"id":50046972,"identity":"f1c6fd11-75f9-4aae-a132-1742d5c60a4f","added_by":"auto","created_at":"2024-01-23 16:06:30","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":355119,"visible":true,"origin":"","legend":"\u003cp\u003eSimulated map of the landscape pattern of the study area in 2020.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/30984ce28ad5bb6b72559fba.jpeg"},{"id":50045738,"identity":"987f0d74-2515-4632-b24f-bd10a5a780cf","added_by":"auto","created_at":"2024-01-23 15:58:30","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":672150,"visible":true,"origin":"","legend":"\u003cp\u003eThe simulated distribution map of the landscape pattern in the study area located in Khorasan Razavi from the years 2040 and 2060.\u003c/p\u003e","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/0ffc2959c11f35c81d8d36a0.jpeg"},{"id":51470845,"identity":"95bb6d46-128d-4946-84b3-591fcbf01f06","added_by":"auto","created_at":"2024-02-22 07:52:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1952134,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3851820/v1/62965f2a-496d-4254-ba07-e73af9332e8f.pdf"}],"financialInterests":"","formattedTitle":"Prediction of Flood-Prone zones based on Cellular Automata in GIS","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eRepetitive natural occurrences like floods lead to notable economic losses and fatalities (Costache \u0026amp; Bui, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Urban regions experiencing robust economic growth bear considerable effects from flood-related risks. Recent environmental shifts (climate change, land use change, etc.), have amplified the risk of floods in cities (Lin et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Pour et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Vulnerability refers to the inherent qualities of urban communities concerning hazards, covering aspects like people, infrastructure, and economic activities, influencing their susceptibility or capacity to manage these risks 'Exposure' describes (Parry et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). It closely relates to the concept of a flooding 'pathway' (DEFRA \u0026amp; EA, 2006), which considers both the location and specific features of urban regions that can either worsen or alleviate the impact of hazards. Risks emerge when urban communities and hazards align spatially, and the severity of these risks depends on the communities' vulnerability, the type of hazard, and the environmental characteristics involved (Lindley et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). However, forecasting flood risks is intricate, involving numerous factors like terrain, river dynamics, precipitation, and land-use (Shafizadeh-Moghadam et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The Cellular Automata (CA) model serves as a spatial simulation tool essential for replicating land use (LU) patterns and configurations ( Gilpin, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). It plays a crucial role in studying the spatial alterations in LU, influenced by both natural phenomena and human factors ( Mor et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eUtilizes a future land-use simulation (FLUS) model to conduct computational evaluations of flood risk based on different scenarios. This FLUS model synergizes an artificial intelligence technique, specifically an artificial neural network (ANN), with a CA model for interactive operations. (Li et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). A pressing need exists to holistically assess various urbanization and dynamic flood risk scenarios concurrently. This research pairs the FLUS and floodplain inundation models how different urban development and climate change scenarios interact. The FLUS model, enhancing precision in simulations, integrates ANN and CA models to replicate nonlinear alterations in land use while factoring in environmental and societal elements (Liu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zhai et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In study (Yuqin et al., 2019), precipitation and LU indicators were analyzed using the (CA) method across the years 1986 to 2010 in the Qinhuai River basin.\u003c/p\u003e \u003cp\u003eThis study introduces a novel space-time framework to predict daily precipitation index at specific spatial extents. These models are integral components of stochastic weather generators, frequently employed in hydrological design or as inputs for environmental and ecosystem models (Richardson, 1981; Wilks \u0026amp; Wilby, 1999; Srikanthan \u0026amp; Wilby, 2001). Efficient examination and surveillance of land cover necessitate extensive data on Earth's surface and habitats. Often, human activities trigger changes in land use and land cover, which are subsequently influenced by natural processes. Conversely, the land use index stands out as a crucial variable determining resource planning and control strategies (Singh et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Various elements regarding the Normalized Difference Vegetation Index (NDVI) and its associated methodologies and procedures have been discussed (Rouse et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1974\u003c/span\u003e). Developing robust methodologies for analyzing NDVI time series data remains a significant challenge in the remote sensing domain. In recent years, several models tailored for modeling NDVI time series have been introduced (Faith, 2015).\u003c/p\u003e \u003cp\u003eThis research creates a simulation technique for LU alterations in the Khorasan Razavi region from 2040 to 2060. Its goal is to construct a unified model chain that incorporates both LU shifts and vulnerability. The amalgamated FLUS chain, spatialization, and flood risk models are suggested to dynamically evaluate forthcoming flood risk and its correlation with societal progress. When tested in the Kashfroud basin, this structure exposes variations in flood risk due to LU changes between 2020 and 2060, providing guidance for future flood prevention strategies amid evolving social circumstances.\u003c/p\u003e"},{"header":"2. Data and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study Area\u003c/h2\u003e \u003cp\u003eKhorasan-Razavi province, situated in northeastern Iran, lies amidst two mountain ranges, Hazar Mosque to the north and Binaloud to the south. The region is crossed by the Kashfroud River, referred to as the Kashfroud catchment area. The study area is positioned between longitudes E58˚30˚-61˚ and latitudes N35˚30˚-37˚ within Khorasan province. The Kashfroud River plays a vital role as the primary water source for the Mashhad Plain, spanning approximately 250 km from its origin to its outlet and comprised of 25 branches. This area experiences predominantly hot and arid weather conditions, with lower altitudes compared to northern Khorasan. The vegetation primarily comprises plants requiring minimal water due to limited rainfall in the region (mousavi \u0026amp; arian, 2015 ; davari et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Given the locations within the Kashfroud River vicinity, the region has seen continuous and seasonal activity throughout various cultural settlements. The total area spans approximately 8996 square kilometers, with 60% comprising mountainous terrain and 40% plains. The Mashhad plain, situated within relatively expansive valleys, extends up 28 to 30 km around Mashhad. Altitudes within the plain range from 890 to 1800 meters above sea level, with the highest point being the peak of the Binaloud mountain range, reaching 3287 meters, and the Hazar Mosque mountain range peak at 3150 meters. The area's lowest point, situated at the outlet, stands at 891 meters above sea level (davari et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Azamirad \u0026amp; Esmaili, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The basin's rivers experience seasonal fluctuations in response to the region's climate, occasionally leading to hazardous floods. Recent droughts and excessive human utilization of water resources beyond the basins' eco-hydrological capacity have created an adverse environmental condition within the Kashfroud watershed (Abedini \u0026amp; Pour Farrash Zadeh, 2021).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Data Resources and Processing Steps\u003c/h2\u003e \u003cp\u003eIn assessing flood risk, the initial step involves building a spatial database encompassing various flood factors. However, the selection of suitable flood factors differs based on specific area characteristics, with these factors exhibiting distinct impacts across different regions. In this study, the Kashafrud Basin's features were considered, drawing from prior research to choose two data aspects: disaster-causing factors and disaster-prone environments. Landsat_8 OLI\u0026thinsp;+\u0026thinsp;TIRS/ETM\u0026thinsp;+\u0026thinsp;remote sensing image data from 2000 and 2020 were collected, while Landsat 7 and 8 remote sensing images were utilized for data interpretation in 2040 and 2060 (Xu et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The analysis utilized LU data as its basis, which entailed the manual correction of raw LU data. Following data correction and manual interpretation, the classification scheme presented by Sciences was implemented, categorizing LU into six monitored classes: residential areas, non-residential areas, vegetation, rivers, mountains, and flood-prone areas (Liu et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). This data model can be accessed using the site \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://eos.com/landviewer/?tool-timelapse\u003c/span\u003e\u003cspan address=\"https://eos.com/landviewer/?tool-timelapse\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The mathematical challenges within diverse remote sensing data analysis techniques can be viewed as distinct instances of input-output data amalgamated with a specific network reliant on algorithms(Zhang et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe Normalized Difference Vegetation Index (NDVI) functions as a standard gauge for generating images that showcase the verdant or relative biomass status of an area. This scale spans from \u0026minus;\u0026thinsp;1.0 to 1.0, predominantly reflecting vegetation. Negative values typically stem from clouds, water, or snow, while values near zero often denote rocks or barren soil. Leveraging the vegetation data within the Landsat imagery, efforts were made to minimize errors when applying the NDVI to Landsat images from 2000 and 2020, aiming to assess LU in urban or urban sprawl contexts (Alsharif et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe primary cause of flood hazards within the surveyed region is primarily attributed to intense rainfall, where higher precipitation indicates an increased risk of potential flooding. Precipitation was selected as a key factor contributing to flood hazards, particularly with a rise in the frequency of heavy rainfall events. In subsequent research, annual precipitation data from 2000 to 2020 was obtained using the website \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://earthengine.google.com/\u003c/span\u003e\u003cspan address=\"https://earthengine.google.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, and the data was downloaded at a resolution of 1*1 km (Wu et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).The Advanced Thermal Reflectance Digital Earth Model (DEM) and Space Emission Radiometer (ASTER) with a 30-meter resolution are used as fundamental data sources for land elevations and slopes (Xu et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The disaster-breeding environments incorporated seven elements: topographical aspects (DEM, slope, hillshade, aspect, flow accumulation, flow direction), along with Euclidean distance, among other spatial factors crucial in evaluating flood risk in Khorasan Razavi (Mohamoud, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Vojtek \u0026amp; Vojtekov\u0026aacute;, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe digital elevation model (DEM) illustrates the elevations within the Kashfroud basin, inversely correlated with flood risk arising from water flow between different elevations. Obtained from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://search.earthdata.nasa.gov/search\u003c/span\u003e\u003cspan address=\"https://search.earthdata.nasa.gov/search\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e for this study, the DEM indicates topographic undulations, with slope significantly influencing flood occurrence (Costache \u0026amp; Bui, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Slope directly impacts surface runoff and vertical percolation, affecting water accumulation, particularly in areas with low slopes that pose higher flood risks. ArcGIS was employed to calculate slope from the DEM dataset. The multiple flow direction algorithm allocates the flow to neighboring downslope cells based on slope, following a weighted distribution (Quinn et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1991\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1995\u003c/span\u003e).Flow accumulation plays a crucial role in grasping the topographic influence on water. The DEM has been instrumental in automatically extracting drainage networks. Employing a fully parallel algorithm, we calculate the comprehensive flow directions ( Do et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The utilization of DEM-based Hillshade enhances the depiction of elevation and topography within the area, offering valuable insights for land slope assessments, identification of peaks and valleys, understanding hydrological systems, and other topographic characteristics. We affirm that producing and precisely interpreting shaded relief images can significantly aid in addressing diverse issues.( Bajabaa et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Topographic elements such as aspect, obtained from field surveys or digital elevation models (DEM), are commonly applied in modeling species distributions. However, extending these models to predict across larger regions or in future scenarios encounters challenges due to the intricate interplay between slope, aspect, and their specific relationships ( Bennie et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). This approach typically generates spatial concentrations resembling distinct lines with increased realism. During floods, riverbanks become inundated, making the distance to the river (DR) crucial for assessing flood risk in specific areas. Calculated using Euclidean distance (Costache et al., 2017), DR is essential for evaluating inundation risks along riverbanks during flooding events.The impact factor data underwent normalization, scaling all values between 0 and 1 for subsequent data mining processes (Muis et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Future LU Simulation (FLUS) Model\u003c/h2\u003e \u003cp\u003eThe FLUS model, a novel simulation framework for LU changes, comprises two primary components: (1) an ANN-based module estimating the probability of occurrence and (2) a (CA) module incorporating a self-adaptive inertial competition mechanism using roulette selection. This model integrates factors like probability of occurrence, neighborhood influence, conversion cost, and inertia coefficient to compute the total conversion probability for each LU type within cells (Liu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).Following this, the roulette mechanism is employed to portray competition among diverse LU types, determining their conversion and distribution on cells for a more accurate simulation of LU changes. This process is typically estimated using Eq.\u0026nbsp;(1) (Liang et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e):\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({TP}_{t(m,n)}={P}_{m,n}\\times {\\varOmega }_{t(m,n)}\\times Inertiat n\\times \\left(1-scc\\to n\\right)\\)\u003c/span\u003e \u003c/span\u003e ( 1 )\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({\\text{T}\\text{P}}_{\\text{t}(\\text{m},\\text{n})}\\)\u003c/span\u003e \u003c/span\u003erefers to the likelihood of grid cell m transitioning from its initial landscape type to the target landscape type n at iteration time t ; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{P}}_{\\text{m},\\text{n}}\\)\u003c/span\u003e\u003c/span\u003e represents the probability of occurrence of landscape type n on grid cell m; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({{\\Omega }}_{\\text{t}(\\text{m},\\text{n})}\\)\u003c/span\u003e\u003c/span\u003e signifies the impact of landscape type n's neighborhood on grid cell m during iteration time t; Inertiat n denotes the inertia coefficient of landscape type n at iteration time t, and scc\u0026rarr;n indicates the conversion cost from the initial landscape type c to the target landscape type n.\u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 ANN\u003c/h2\u003e \u003cp\u003eThe simulation primarily assesses the likelihood of transforming various land-use categories. This process employs a neural network model comprising three layers: input, hidden, and output layers. By randomly sampling data from diverse spatial driving factors and multiple-factor urban land-use distribution layers, the artificial neural network (ANN) is trained to determine the probability of converting different land-use types, effectively estimating the suitability probability for expanding cell use. The operational functionality of the network as depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. ( Cai et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe initial module selected 11 influential factors to establish the Artificial Neural Network (ANN) model for the probability of occurrence concerning each LU type. The chosen factors encompassed both natural aspects like DEM, slope, hillshade, flow accumulation, NDVI, flow direction, and aspect, along with precipitation and distance metrics such as distance to rivers or Euclidean Distance. ANN, a model inspired by biological neural network simulations, handles complex nonlinear functions with numerous variables. Through iterative learning and recall processes, it fits intricate relationships between input data and training objectives, ensuring the generation of more accurate suitability probability distributions. This process establishes the correlation between the probability of each land type and the associated driving factors, expressed in Eq.\u0026nbsp;(2) as: The suitability probability calculation is represented in the equation as:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({p}_{m,n }=p\\left(m,n,t\\right)= \\sum _{i}{w}_{i,n }\\times sigmoid\\left({net}_{i}\\left(m,t\\right)\\right)= \\sum _{i}{w}_{i,n} \\times \\frac{1}{1+ {e}^{{net}_{i}\\left(m,t\\right)}}\\)\u003c/span\u003e \u003c/span\u003e ( 2 )\u003c/p\u003e \u003cp\u003eThis equation defines p(m, n, t) as the suitability probability of landscape type non-grid cell m at time t. Here, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{w}}_{\\text{i},\\text{n} }\\)\u003c/span\u003e\u003c/span\u003e and sigmoid denote the weight and activation function between the hidden layer and the output layer, respectively. Additionally, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{n}\\text{e}\\text{t}}_{\\text{i}}\\left(\\text{m},\\text{t}\\right)\\)\u003c/span\u003e\u003c/span\u003e represents the signal received from grid m at time t in the hidden layer (Liu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 CA\u0026ndash;Markov Model\u003c/h2\u003e \u003cp\u003eThe CA model functions as a spatial dynamic model with discrete attributes in time, space, and states, integrating local spatial interactions and temporal cause-and-effect relationships ( Wolfram, 1983) ,Operating on a structured grid system, each cell possesses a finite set of discrete states and neighboring cells, where its future state is determined by its own state and the states of neighboring cells in the previous time step. Employing rules governing cell state changes enables the simulation of spatial and temporal changes in LU and various intricate systems (Li \u0026amp; yeh, 2002).The model representation is depicted as follows ( Wolfram, 1983):\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(S\\left(t+1\\right)=F\\left(S\\left(t\\right),N\\right)\\)\u003c/span\u003e \u003c/span\u003e ( 3 )\u003c/p\u003e \u003cp\u003eS(t) and S(t\u0026thinsp;+\u0026thinsp;1) represent the collection of cell states at times t and t\u0026thinsp;+\u0026thinsp;1 correspondingly; F denotes the regulation governing cellular state transitions, while N signifies the neighborhood filter. The Markov model stands as a classical statistical and quantitative forecasting technique ( Wijesekara et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), Developed by Andrey Markov from the former Soviet Union, this method finds application across various fields like natural language processing, human resource management, and LU simulation, among others.\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(S\\left(t+1\\right)=P\\times S\\left(t\\right)\\)\u003c/span\u003e \u003c/span\u003e ( 4 )\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(P=\\left[\\begin{array}{c}{p}_{11} {p}_{12} \\dots {p}_{1n}\\\\ {p}_{21} {p}_{22} \\dots {p}_{2n}\\\\ . . . .\\\\ . . . .\\\\ . . . .\\\\ {p}_{n1} {p}_{n2} \\dots {p}_{nn}\\end{array}\\right]\\)\u003c/span\u003e \u003c/span\u003e ( 5 )\u003c/p\u003e \u003cp\u003ewhere S(t) and S(t\u0026thinsp;+\u0026thinsp;1) are the LU states at times t and t\u0026thinsp;+\u0026thinsp;1, respectively; P is the state transition probability matrix; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({p}_{ij}\\)\u003c/span\u003e\u003c/span\u003e is the probability of class i land transforming into\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\text{c}\\text{l}\\text{a}\\text{s}\\text{s} \\text{j} \\text{l}\\text{a}\\text{n}\\text{d},{\\text{p}}_{\\text{i}\\text{j}}\\in \\left(\\text{0,1}\\right),\\text{a}\\text{n}\\text{d}\\sum _{\\text{j}=1}^{\\text{n}}{\\text{p}}_{\\text{i}\\text{j}}=1,\\left(\\text{i},\\text{j}=\\text{1,2},3\\dots ,\\text{n}\\right). \\left(6\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eIn LU simulation, the Markov model concentrates on predicting change quantities but lacks the ability to portray the spatial distribution or patterns of various land changes ( Nor et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) Conversely, the CA model forecasts the spatial-temporal characteristics of diverse land changes, addressing the Markov model's limitations ( Wijesekara et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The model is depicted as follows:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$${\\text{s}}_{\\text{i}\\text{j}}\\left(\\text{t}+1\\right)=\\text{F}({\\text{s}}_{\\text{i}\\text{j}}\\left(\\text{t}\\right),{\\text{Q}}_{\\text{i}\\text{j}}\\left(\\text{t}\\right),\\text{V})$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e7\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({s}_{ij}\\)\u003c/span\u003e \u003c/span\u003e(t) and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({s}_{ij}\\)\u003c/span\u003e\u003c/span\u003e (t\u0026thinsp;+\u0026thinsp;1) represent the conditions of the cell in row i and column j at times t and t\u0026thinsp;+\u0026thinsp;1, respectively; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Q}_{ij}\\)\u003c/span\u003e\u003c/span\u003e(t) denotes the state of the neighboring cells of the cell in row i and column j at time t; V represents the suitability atlas; and F is the rule governing cell transformations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 Kappa Coefficient\u003c/h2\u003e \u003cp\u003eThe Kappa coefficient is commonly employed to assess the precision of the anticipated LU map. Typically, a Kappa value surpassing 0.75 is considered indicative of superior accuracy in the prediction outcome ( Araya \u0026amp; cabral, 2010 ; Tong et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). If the total number of pixels is N, and S represents the count of accurately simulated grids, where the actual counts for each class are a1, a2,. . ., ac, and the simulated counts for each class are b1, b2,. . ., bc, respectively, the Kappa coefficient (K) can be mathematically represented using the following formulas:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\text{K}=\\frac{{\\text{P}}_{0}-{\\text{P}}_{\\text{c}}}{1-{\\text{P}}_{\\text{c}}} \\left(8\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$${\\text{P}}_{0}=\\frac{\\text{S}}{\\text{N}} \\left(9\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$${\\text{P}}_{\\text{c}}=\\frac{{\\text{a}}_{1}\\times {\\text{b}}_{1}\\times {\\text{a}}_{2}\\times {\\text{b}}_{2}+\\dots +{\\text{a}}_{\\text{c}}\\times {\\text{b}}_{\\text{c}}}{\\text{N}\\times \\text{N}} \\left(10\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe general accuracy measures the proportion of correct predictions made by the model across all test sets against the total count. The model exhibits high accuracy when Kappa is \u0026le;\u0026thinsp;0.75 and an overall accuracy (OA) is less than 1, moderate accuracy when Kappa is \u0026le;\u0026thinsp;0.5 and OA is less than 0.75, and poor accuracy when Kappa is \u0026le;\u0026thinsp;0 and OA is less than 0.5.\u003c/p\u003e \u003cp\u003eurthermore, the validation of the alterations was confirmed through the employment of the figure of merit (Fom), considered more effective than the Kappa coefficient in evaluating the precision of simulated alterations (Pontius et al. 2008).\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\text{F}\\text{o}\\text{m}=\\frac{\\text{B}}{\\text{A}+\\text{B}+\\text{C}+\\text{D}}\\times 100\\text{%} \\left(11\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHere, A denotes an error area resulting from observed change predicted as persistence, B signifies an accurate area resulting from observed change predicted as change, C represents an error area resulting from observed change predicted as shifting to an incorrect category, and D indicates an error area resulting from observed persistence predicted as change.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3. Results and Analysis","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Emporal Alterations in Land Utilization\u003c/h2\u003e \u003cp\u003eThe forecast of future LU demand was conducted using the Markov chain technique.The pixel-based data has a resolution of 30 m. ArcGIS 10.5 facilitated the reclassification and graphical representation of LU maps in Khorasan Razavi for 2000 and 2020, The Markov chain module was employed to compute the LU requirements for different types in the years 2040 and 2060. (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). To assess the demand for various LU types in 2040 and 2060, Lu data from 2000 and 2020 were incorporated into the Markov chain module. Details regarding the area of each LU and its evolving trend are outlined in (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).The land types were mainly classified into established, comprising residential areas, non-residential areas, vegetation, rivers, mountains, and flood-prone areas. The area changes of the six types of LU for the towenty years were also counted.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eLU classes distribution in area (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{k}\\text{m}}^{2}\\)\u003c/span\u003e\u003c/span\u003e) in khorasan razavi from 2000 and 2020.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResidential\u003c/p\u003e \u003cp\u003earea\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon residential area\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMountain\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRiver\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAccidental\u003c/p\u003e \u003cp\u003espots\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e288.99 15.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e489.65 26.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e190.81 10.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e795.83 42.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e83.21 4.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.54 0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e316.35 17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e469.68 25.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e152.46 8.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e818.62 44.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e91.01 4.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.05 0.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eLU data from 2000 and 2020 were used to determine the required parameters of the FLUS model. A significant expansion in residential areas, vegetation, accident-prone spots and Razavi rivers of Khorasan was observed when analyzing changes in latitude and time (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). On the contrary, during this period, a decrease in the area covered by mountainous and non-residential areas was observed. These changes between 2000 and 2020 significantly affected the displacement and strengthening of vulnerable areas.\u003c/p\u003e \u003cp\u003eTo train the ANN model, 2056890 samples were randomly selected in the study area. To evaluate the accuracy of the FLUS model, it was used with a random sample, which resulted in a figure of merit (FOM) of 0.064 and a kappa coefficient of 0.94 for simulation. Despite the relatively low FoM, previous LU modeling experiments typically vary the reported values in the range of 0.1\u0026ndash;0.3. This inconsistency is caused by the effects of path dependence, which causes inaccuracy in the prediction of land change models. Based on LU data from 2000 and 2020, the demand for different types of LU in 2040 and 2060 was predicted using the Markov chain modul.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Drivers of Land-Use Change\u003c/h2\u003e \u003cp\u003eThe selection of driving factors hinges on research aims, regional traits, and data accessibility.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe generation of suitability maps and the LU conversion probability matrix, as well as the transition suitability maps, heavily relies on these factors and limited areas. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e showcases spatial analyses concerning natural factors influencing urban land expansion. For this study's context, natural driving factors comprised DEM, slope, slope direction, aspect, hillshade, and annual precipitation. Additionally,Euclidean distance analysis is for rivers, flow accumulation, flow direction locations, and NDVI, regarded as ecological factors impacting LU transformation. These constituted the primary drivers of LU changes within this research. Spatial data derived from diverse sources, including geocoded location coordinates, underwent manipulations such as clipping and merging to align with the coverage area. Nine natural factors, delineated in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, served as the driving forces of change. These factors were defined based on the researchers' expertise in the study area, with weights derived from a series of paired comparisons.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Conversion Expenses and Limitations on Area Changes\u003c/h2\u003e \u003cp\u003eConversion cost is utilized to quantify the level of complexity associated with transitioning from the current LU type to the intended type, playing a significant role in molding LU dynamics. Various factors influence the conversion processes between different LU types, typically including: (a) probability of occurrence, and (b) LU pair conversion cost, which signifies the challenge in shifting from one LU and Land Cover (LULC) class to another. This serves as a mathematical representation of the Earth's motion direction. In the cost matrix settings, a value of 0 denotes a higher land conversion cost and diminished potential for land expansion, while a value of 1 signifies a lower land conversion cost and higher potential for land expansion. Referring to the LU transition matrix from 2000 to 2020, the final cost matrix for LU change was established (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eConversion cost coefficients between LU types.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResidential\u003c/p\u003e \u003cp\u003earea\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon residential area\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003evegetation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003emountain\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eriver\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAccidental\u003c/p\u003e \u003cp\u003espots\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResidential\u003c/p\u003e \u003cp\u003earea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNON residential area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003evegetation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003emountain\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eriver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAccidental\u003c/p\u003e \u003cp\u003espots\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Neighborhood Factors\u003c/h2\u003e \u003cp\u003eNeighborhood factors indicate the capacity for different land types to expand, particularly their potential expansion under external influences. Ranging between 0 and 1, higher values signify a greater ability for a specific LU type to expand. These factors are determined by scrutinizing the LU data within the study area and integrating expert insights. Additionally, by scrutinizing the LU conversion matrix from 2000 to 2020 within the basin, the FLUS model's neighborhood weights were derived. Initial computations were based on the proportion of the extended area for each LU type, with continuous optimization and adjustment during the simulation process to obtain the optimal neighborhood weight parameters for the experiment, as depicted in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eweight of Neighborhood parametrs.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand use types\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResidential\u003c/p\u003e \u003cp\u003earea\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon residential area\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMountain\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRiver\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAccidental\u003c/p\u003e \u003cp\u003espots\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNeighborhood weight\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Simulation of LU Arrangement through FLUS Modeling\u003c/h2\u003e \u003cp\u003eThis study utilized ArcGIS 10.5 to convert spatial data and standardized the data of Khorasan Razavi province, maintaining a resolution of 30 meters as the standard. Processing was conducted employing the GeoSOS-FLUS V2.4 model to simulate and forecast flood hotspots and their spatial fluctuations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Calculation of Suitability Probability using Neural Network Techniques\u003c/h2\u003e \u003cp\u003eFor this investigation, precipitation, NDVI, DEM, slope, direction, Euclidean distance, hill shadow, direct flow, and area flow accumulation were extracted as influential factors for floods and their locations. During the artificial neural network training with GeoSOS-FLUSV2.4, a uniform distribution sampling approach with a sampling ratio of 1% was adopted, configuring the hidden layer to consist of 13 nodes. Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e displays the probability maps generated for each factor. The calculated root mean square error (RMSE) during training was 0.21, signifying a high level of training accuracy. The RMSE formula is as follows:\u003cdiv id=\"Equf\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equf\" name=\"EquationSource\"\u003e\n$$\\text{R}\\text{M}\\text{S}\\text{E}=\\sqrt{\\frac{1}{\\text{n}} \\sum _{\\text{i}=1}^{\\text{n}}{\\left(\\widehat{{\\text{y}}_{\\text{i}}}-{\\text{y}}_{\\text{i}}\\right)}^{2}} \\left(12\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere n is the number of data points, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\widehat{{\\text{y}}_{\\text{i}}}\\)\u003c/span\u003e\u003c/span\u003eis the predicted value of the data point, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{y}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003eis the observed value of the data point.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThis study predicted the future changes in different types of LU by analyzing the classification data of the initial and final years and used the SD model for prediction. The relative influence of neighborhood factors was obtained from multiple tests, considering conversion rates among different LU types in the confusion matrix. After the finalization of the simulation, the output shows the results of the simulation. In addition, an estimation model using ANN was built based on the previously mentioned driving forces, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. This model effectively shows the probability of occurrence for each type of LU.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Validity of the Model\u003c/h2\u003e \u003cp\u003eValidation of the accuracy of FLUS model in predicting future exploitation in Razavi Khorasan province requires the verification of the previous model. Using LU data from 2000 as the initial dataset, the FLUS model simulated the 2020 LU pattern. The spatial image of simulated LU in 2020 in the study area is presented in (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The findings showed a strong alignment between actual and simulated LU patterns for 2020. Comparative analysis between real and simulated user maps enabled calculation of Kappa coefficient and FoM. The calculated values for the 2020 simulation results show a Kappa coefficient of 0.94 and a FoM of 0.064, indicating a satisfactory level of accuracy in the simulation. The accuracy assessment of the 2020 landscape pattern simulation included visual interpretations that showed a significant similarity between the predicted 2020 map and the actual map. Further visual analysis of the zoomed-in simulation results from three specific areas strengthened the FLUS model's ability to accurately predict future LU in the study area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.8 Change of Future Spatial-temporal Use in Khorasan Razavi\u003c/h2\u003e \u003cp\u003eBy interpreting remote sensing images from Landsat 7 ETM\u0026thinsp;+\u0026thinsp;and Landsat-8 OLI TIRS and cross-verifying and rectifying the outcomes, the recent LU maps of the study zone for both 2000 and 2020, meeting the stipulated accuracy criteria (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn summary, vulnerable areas are projected to become larger, with an expected increase of 3.47 km2 by 2040 and 3.85 km2 by 2060, representing approximately 0.18% and 0.20% of the total area, respectively.\u003c/p\u003e \u003cp\u003eUsing the simulated model for LU change, based on the 2020 LU data as basic data and using the probability of each type of LU as automatic cell conversion rules, forecast maps for 2040 and 2060 were prepared. The maps show the areas under the natural development scenarios shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThese findings indicate a gradual expansion of flood-prone areas between 2040 and 2060 under the natural development scenario and we see more illegal constructions and agricultural lands in the river.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe amount of changes relative to the LU area in the study area in the future from 2020 to 2060 under the natural scenario.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResidential\u003c/p\u003e \u003cp\u003earea\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon residential area\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMountain\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRiver\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAccidental\u003c/p\u003e \u003cp\u003espots\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({km}^{2}\\)\u003c/span\u003e\u003c/span\u003e %\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020 S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e316.35 17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e469.68 25.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e152.44 8.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e818.50 44.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e91.00 4.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.05 0.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2040 S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e328.99 17.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e461.01 24.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e137.20 7.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e821.73 44.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e98.64 5.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.47 0.18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2060 S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e334.78 18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e456.54 24.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e130.99 7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e818.69 44.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e106.2 5.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.85 0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eInvestigating the interaction between future urbanization trends and environmental transformations, including the increase of disaster prone areas due to climate change (rainfall) and changes in river boundaries, is very important to prevent unforeseen disasters. The relationship between these elements is complex and constantly changing. In our current review, the evaluation relied on a representative computing scenario. The method is robust and comprehensive, adept at handling the complexities of the factors involved, and proves to be an effective approach for determining the most accurate locations. Such identification serves as a basis for innovative planning and strategy aimed at reducing flood risk.\u003c/p\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Limitations and Future works\u003c/h2\u003e \u003cp\u003eThere are constraints within our research that necessitate refinement in future endeavors. Initially, there is ample room for enhancing the model's predictive precision. This study evaluates the performance of the FLUS model through Kappa and OA metrics, exhibiting a commendable level of predictive accuracy. Moreover, it establishes that 11 influential factors contribute to modeling and forecasting accident-prone areas in Khorasan Razavi. Integrating human and natural driving factors into LU change through the FLUS model, coupled with ANN, can yield more pragmatic simulation outcomes. However, the accuracy of predictions could be heightened by incorporating the latest land cover data. On one hand, this study is founded on DEM data modeling, which might result in underestimation or overestimation of simulation efficacy due to DEM data inaccuracies. On the other hand, amalgamating rainfall and land vegetation data bolsters the dependability of severe flood predictions. The conversion of other LUs into urban utilization substantially escalates flood hazards owing to alterations in impermeable surfaces. Consequently, in forthcoming forecasts, it is crucial to dynamically adjust pertinent factors influencing peak flows and flood risk to bolster forecast accuracy. Despite the inevitability of uncertainties in flood risk assessments, the divergence in outcomes produced by the proposed model remains within an acceptable range, ensuring the accuracy of flood risk evaluations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.2.Suggestions regarding Strategies and Policies for Urban Flooding\u003c/h2\u003e \u003cp\u003eIn the 21st century, urban flooding poses a significant challenge for cities. Human responses to flood impacts heavily rely on how urban facilities are allocated and how managers plan for the future. The recent flooding in Khorasan Razavi province inflicted severe damage on both infrastructure and natural resources. These floods resulted in road destruction, loss of agricultural fields, and damage to residences and structures. The repercussions of these natural calamities resulted not just in financial setbacks but also in human, psychological, and societal losses. The enduring impact of these floods on people's lives and the rehabilitation of affected areas have presented numerous challenges for the local community. Hence, there's an urgent call for implementing flood risk adaptation strategies in Khorasan Razavi.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study introduces an approach using a real options strategy for urban planning and strengthening vulnerable sites against flood risks in the midst of environmental change. Various evaluations including overall accuracy, kappa coefficient and figure of merit (Fom) were performed in different programs. The simulated results for future LU show high simulation accuracy that aligns with research requirements. Considering the impact of urbanization and environmental changes, the model provides a significant insight into the extent and spatial expansion of flood hazards in Khorasan Razavi, which indicates a significant intensification of flooded urban areas in the future.\u003c/p\u003e \u003cp\u003eThe simulations predict an increase in the built-up area along the river banks in 2040 and 2060 compared to 2020. It is estimated that these constructions along the river will cover approximately 5.3 to 5.7% of the entire river due to the increase in water volume due to environmental changes. to increase by 3.85 square kilometers compared to 2020 and provide insights into flood risk areas. determined by simulation. This analysis deals with the spatial distribution of urban and rural areas prone to future floods and enables a comprehensive understanding of urban landscapes and challenges from the perspective of adaptive urban planning. It prioritizes flood mitigation strategies and highlights critical areas in the region.\u003c/p\u003e \u003cp\u003eThis paper introduces a new framework using the real options approach for planning and designing urban flood control measures and addresses the impact of environmental changes on urban flooding. However, there are areas for further improvement in this study. Considering the uncertainty in precipitation forecast due to global climate model accuracy, the use of dynamic downscaling methods can increase future precipitation forecasts. In addition, future research could explore additional measures for urban resilience to combat the effects of environmental change on urban flooding.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eWe extend our appreciation to the Water Management Institute of Iran for granting access to the flood database, facilitating data acquisition for this research.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eCOI Statement\u003c/h2\u003e\n\u003cp\u003eOn behalf of all authors, the corresponding author states that there is no conflict of interest.\u003c/p\u003e\n\u003ch2\u003eDeclaration of Interests\u003c/h2\u003e\n\u003cp\u003eThe authors have no relevant financial or non financial intest to disclose.\u003c/p\u003e\n\u003ch2\u003e\u0026nbsp;Data Availibility\u003c/h2\u003e\n\u003cp\u003eData and programs will be made available on request\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbedini M, Zadeh PF, F (2021) Analysis and Modeling of the Relationship between Monthly Discharge and Geomorphometric Characteristics (Case Study: Kashafrood Watershed). Geogr Environ Plann 32(4):29\u0026ndash;44\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlsharif M, Alzandi AA, Shrahily R, Mobarak B (2022) Region Saudi Arabia Forests 13(10):1530Land Use Land Cover Change Analysis for Urban Growth Prediction Using Landsat Satellite Data and Markov Chain Model for Al Baha\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAraya YH, Cabral P (2010) Analysis and modeling of urban land cover change in Set\u0026uacute;bal and Sesimbra, Portugal. Remote Sens 2(6):1549\u0026ndash;1563\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArora A, Arabameri A, Pandey M, Siddiqui MA, Shukla UK, Bui DT, Bhardwaj A (2021) Optimization of state-of-the-art fuzzy-metaheuristic ANFIS-based machine learning models for flood susceptibility prediction mapping in the Middle Ganga Plain, India. Sci Total Environ 750:141565\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAzamirad M, Esmaili K (2018) Investigation flooding potential in the Kashafrud watershed, Mashhad the method SCS and GIS. J Watershed Manage Res 9(17):26\u0026ndash;38\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBajabaa S, Masoud M, Al-Amri N (2014) Flash flood hazard mapping based on quantitative hydrology, geomorphology and GIS techniques (case study of Wadi Al Lith, Saudi Arabia). Arab J Geosci 7:2469\u0026ndash;2481\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBennie J, Huntley B, Wiltshire A, Hill MO, Baxter R (2008) Slope, aspect and climate: spatially explicit and implicit models of topographic microclimate in chalk grassland. Ecol Model 216(1):47\u0026ndash;59\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai Y, Zong W, Jiao S, Wang Z, Ou L (2023) Land-Use Assessment and Trend Simulation from a Resilient Urban Perspective: A Case Study of Changsha City. Sustainability 15(18):13890\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChapi K, Singh VP, Shirzadi A, Shahabi H, Bui DT, Pham BT, Khosravi K (2017) A novel hybrid artificial intelligence approach for flood susceptibility assessment, vol 95. Environmental modelling \u0026amp; software, pp 229\u0026ndash;245\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCostache R, Bui DT (2019) Spatial prediction of flood potential using new ensembles of bivariate statistics and artificial intelligence: A case study at the Putna river catchment of Romania. Sci Total Environ 691:1098\u0026ndash;1118\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCostache R, Bui DT (2020) Identification of areas prone to flash-flood phenomena using multiple-criteria decision-making, bivariate statistics, machine learning and their ensembles. Sci Total Environ 712:136492\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCostache R, Pham QB, Avand M, Linh NTT, Vojtek M, Vojtekov\u0026aacute; J, Dung TD (2020) Novel hybrid models between bivariate statistics, artificial neural networks and boosting algorithms for flood susceptibility assessment. J Environ Manage 265:110485\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavari MS, Basafa H, Toghraei M (2020) Kashafrud River (Mashhad plain) from the Neolithic to the Middle Bronze Age. J Archaeol Stud 12(3):41\u0026ndash;63\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavari M, Mousivand F, Basafa H (2018) Explanation of Early Pottery Features in Kashafrud River Basin: Based on the typology of Pottery, s Esmail-Abad site in Mashhad. J Archaeol Stud 10(1):75\u0026ndash;94\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDEFRA, \u0026amp; EA (2006) R\u0026amp;D outputs: Flood risks to people. Phase 2. FD2321/TR1 The flood risks to people methodology. Department for Environment Food and Rural Affairs and the Environment Agency, London\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDo HT, Limet S, Melin E (2011) Parallel computing flow accumulation in large digital elevation models. Procedia Comput Sci 4:2277\u0026ndash;2286\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao Y, Chen J, Luo H, Wang H (2020) Prediction of hydrological responses to land use change. Sci Total Environ 708:134998\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGilpin W (2019) Cellular automata as convolutional neural networks. Phys Rev E 100(3):032402\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Yeh AGO (2002) Neural-network-based cellular automata for simulating multiple land use changes using GIS. Int J Geogr Inf Sci 16(4):323\u0026ndash;343\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Chen G, Liu X, Liang X, Wang S, Chen Y, Xu X (2017) A new global land-use and land-cover change product at a 1-km resolution for 2010 to 2100 based on human\u0026ndash;environment interactions. Annals of the American Association of Geographers 107(5):1040\u0026ndash;1059\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang X, Liu X, Li X, Chen Y, Tian H, Yao Y (2018) Delineating multi-scenario urban growth boundaries with a CA-based FLUS model and morphological method. Landsc urban Plann 177:47\u0026ndash;63\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin J, He P, Yang L, He X, Lu S, Liu D (2022) Predicting future urban waterlogging-prone areas by coupling the maximum entropy and FLUS model. Sustainable Cities and Society 80:103812\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindley SJ, Handley JF, Theuray N, Peet E, McEvoy D (2006) Adaptation strategies for climate change in the urban environment: assessing climate change related risk in UK urban areas. J Risk Res 9(5):543\u0026ndash;568\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu J, Liu M, Zhuang D, Zhang Z, Deng X (2003) Study on spatial pattern of land-use change in China during 1995\u0026ndash;2000. Sci China Ser D: Earth Sci 46:373\u0026ndash;384\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu J, Wang J, Xiong J, Cheng W, Cui X, He W, Wang N (2021) Dynamic Assessment of the flood risk at basin scale under simulation of land-use scenarios and spatialization technology of factor. Water 13(22):3239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu X, Liang X, Li X, Xu X, Ou J, Chen Y, Pei F (2017) A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects. Landsc Urban Plann 168:94\u0026ndash;116\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohamoud YM (1992) Evaluating Manning's roughness coefficients for tilled soils. J Hydrol 135(1\u0026ndash;4):143\u0026ndash;156\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMor B, Garhwal S, Kumar A (2021) A systematic review of hidden Markov models and their applications. Arch Comput Methods Eng 28:1429\u0026ndash;1448\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMosavi E, Arian M (2015) Neotectonics of Kashaf Rud River, NE Iran by Modified Index of Active Tectonics (MIAT). Int J Geosci 6(07):776\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuis S, Verlaan M, Winsemius HC, Aerts JC, Ward PJ (2016) A global reanalysis of storm surges and extreme sea levels. Nat Commun 7(1):11969\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuriithi FK (2015) Centered log-ratio (clr) transformation and robust principal component analysis of long-term NDVI data reveal vegetation activity linked to climate processes. Climate 3(1):135\u0026ndash;149\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNor ANM, Corstanje R, Harris JA, Brewer T (2017) Impact of rapid urban expansion on green space structure. Ecol Ind 81:274\u0026ndash;284\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParry ML, Canziani OF, Palutikof JP, Linden PJ, Hanson CE (2007) IPCC Fourth Assessment Report: Climate Change 2007. The Fourth Assessment Report of the Intergovernmental Panel on Climate Change\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePour SH, Abd Wahab AK, Shahid S, Asaduzzaman M, Dewan A (2020) Low impact development techniques to mitigate the impacts of climate-change-induced urban floods: Current trends, issues and challenges. Sustainable Cities and Society 62:102373\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuinn PFBJ, Beven K, Chevallier P, Planchon O (1991) The prediction of hillslope flow paths for distributed hydrological modelling using digital terrain models. Hydrol Process 5(1):59\u0026ndash;79\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuinn PF, Beven KJ, Lamb R (1995) The in (a/tan/β) index: How to calculate it and how to use it within the topmodel framework. Hydrol Process 9(2):161\u0026ndash;182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichardson CW (1981) Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resour Res 17(1):182\u0026ndash;190\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRouse JW, Haas RH, Schell JA, Deering DW (1974) Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec Publ 351(1):309\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShafizadeh-Moghadam H, Valavi R, Shahabi H, Chapi K, Shirzadi A (2018) Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping. J Environ Manage 217:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh SK, Mustak S, Srivastava PK, Szab\u0026oacute; S, Islam T (2015) Predicting spatial and decadal LULC changes through cellular automata Markov chain models using earth observation datasets and geo-information. Environ Processes 2:61\u0026ndash;78\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSrikanthan R, McMahon TA (2001) Stochastic generation of annual, monthly and daily climate data: A review. Hydrol Earth Syst Sci 5(4):653\u0026ndash;670\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTong ST, Sun Y, Ranatunga T, He J, Yang YJ (2012) Predicting plausible impacts of sets of climate and land use change scenarios on water resources. Appl Geogr 32(2):477\u0026ndash;489\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVojtek M, Vojtekov\u0026aacute; J (2019) Flood susceptibility mapping on a national scale in Slovakia using the analytical hierarchy process. Water 11(2):364\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWijesekara GN, Gupta A, Valeo C, Hasbani JG, Qiao Y, Delaney P, Marceau DJ (2012) Assessing the impact of future land-use changes on hydrological processes in the Elbow River watershed in southern Alberta. Can J hydrology 412:220\u0026ndash;232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilks DS, Wilby RL (1999) The weather generation game: a review of stochastic weather models. Prog Phys Geogr 23(3):329\u0026ndash;357\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolfram S (1983) Statistical mechanics of cellular automata. Rev Mod Phys 55(3):601\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu X, Shen X, Li J, Xie X (2024) Determination and projection of flood risk based on multi-criteria decision analysis (MCDA) combining with CA-Markov model in Zhejiang Province, China, vol 53. Urban Climate, p 101769\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu XL, Liu JY, Zhang ZX, Zhou WC, Zhang SW, Li RD, Shi XZ (2017) A Time Series Land Ecosystem Classification Dataset of China in Five-Year Increments (1990. Journal of Global Change Data \u0026amp; Discovery2017, 1(1), 52\u0026ndash;59\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai Y, Yao Y, Guan Q, Liang X, Li X, Pan Y, Zhou J (2020) Simulating urban land use change by integrating a convolutional neural network with vector-based cellular automata. Int J Geogr Inf Sci 34(7):1475\u0026ndash;1499\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L, Zhang L, Du B (2016) Deep learning for remote sensing data: A technical tutorial on the state of the art. IEEE Geoscience and remote sensing magazine 4(2):22\u0026ndash;40\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Land use (LU) simulation, Cellular automata (CA), Markov chains, ANN, FLUS model","lastPublishedDoi":"10.21203/rs.3.rs-3851820/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3851820/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDue to climate change and rapid urbanization, urban flooding is on the rise, necessitating effective flood control measures in urban areas. Predicting potential flood-prone areas undergoing Land Use (LU) changes could significantly aid in planning for risk reduction and sustainable urban design. However, there's a scarcity of studies that consider both climate change and LU alterations. This research introduces a novel basin-scale framework utilizing a Future LU Simulation (FLUS) model to evaluate disaster-prone areas' risk from 20-year flood scenarios projected for 2040 and 2060. The Markov-FLUS model was developed and validated using historical data from 2000 to 2020. This model was then employed to simulate LU changes from 2020 to 2060 based on natural scenarios. Focusing on Khorasan Razavi as a case study, it investigates the potential consequences of LU transformations due to ongoing urbanization and vegetation changes in connection with predicted environmental shifts. The findings indicate an anticipated increase in accident-prone areas and constructed land in the studied area in the future. Spatially, this heightened flood risk primarily occurs on the periphery of existing developed areas or converted land. This framework's insights into future flood-prone areas' spatio-temporal characteristics offer valuable guidance for implementing rational flood mitigation measures in the most critical regions for development.\u003c/p\u003e","manuscriptTitle":"Prediction of Flood-Prone zones based on Cellular Automata in GIS","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-23 15:58:25","doi":"10.21203/rs.3.rs-3851820/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"10c89403-58ba-4759-bba3-86c880693251","owner":[],"postedDate":"January 23rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-02-22T07:44:37+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-23 15:58:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3851820","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3851820","identity":"rs-3851820","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}