Simulating the urban growth dimensions and scenario prediction through sleuth model: a case study of Rasht County,Guilan, Iran

Urban growth models (UGM) as regional planning tools are of great interest for quantitative analysis of urban complex systems. In this study, the SLEUTH UGM has been calibrated through a sequential multistage automated method to derive the pattern of urban growth in Rasht County from 1975 up to year 2011. Evaluation of model goodness of fit confirms that the model is adjusted properly to the area under investigation. Four growth rules of spontaneous, new spreading center, edge and road influenced growth as well as five coefficients of diffusion, breed, spread, road gravity and slope resistance are responsible to detect quantitative aspects of urban dynamics from control years. According to the results, successive improvement of the model parameters during the calibration mode indicates applicability of the model for forecasting of future urban growth mechanism until the year 2050. Accordingly, two growth scenarios were developed mainly with the aim of investigating the coefficients’ role in controlling the nature of urban dynamics. In this concern, the spread and road gravity coefficients’ value, as two major driving forces of urban sprawl in the study area were reduced to dictate compact and infill growth, compared to their original values derived from calibration for historical prediction. Comparison between two forecasted scenarios indicates insignificant difference in total amount of the urban area, which denotes there is a threshold to urbanization and the current trend of urban growth could not be maintained. Finally, we conclude that Rasht County with considerable industrial and agricultural attractions, will witness noticeable expansion from 20,310 ha in 2011, up to 34,745 ha in 2050, accounting to 71 % increase in total area of manmade surfaces.     

Precipitation pattern modeling using cross-station perception: regional investigation

Establishing robust models for predicting precipitation processes can yield a significant aspect for many applications in water resource engineering and environmental prospective. In particular, understanding precipitation phenomena is crucial for managing the effects of flooding in watersheds. In this research, a regional precipitation pattern modeling was undertaken using three intelligent predictive models incorporating artificial neural network (ANN), support vector machine (SVM) and random forest (RF) methods. The modeling was carried out using monthly time scale precipitation information in a semi-arid environment located in Iraq. Twenty weather stations covering the entire region were used to construct the predictive models. At the initial stage, the region was divided into three climatic districts based on documented research. Initially, modeling was carried out for each district using historical information from regionally distributed meteorological stations for calibration. Subsequently, cross-station modeling was undertaken for each district using precipitation data from other districts. The study demonstrated that cross-station modeling was an effective means of predicting the spatial distribution of precipitation in watersheds with limited meteorological data.     

Performance evaluation of land change simulation models using landscape metrics

This study proposes a landscape metrics-based method for model performance evaluation of land change simulation models. To quantify model performance at both landscape and class levels, a set of composition- and configuration-based metrics including number of patches, class area, landscape shape index, mean patch area and mean Euclidean nearest neighbour distance were employed. These landscape metrics provided detailed information on simulation success of a cellular automata-Markov chain (CA-Markov) model standpoint of spatial arrangement of the simulated map versus the corresponding reference layer. As a measure of model simulation success, mean relative error (MRE) of the metrics was calculated. At both landscape and class levels, the MRE values were accounted for 22.73 and 10.2%, respectively, which are further categorised into qualitative measurements of model simulation performance for simple and quick comparison of the results. Findings of the present study depict a hierarchical and multi spatial level assessment of model performance.