Relative Operating Characteristic Research Articles

Land-Cover Change Analysis and Simulation in Conakry (Guinea), Using Hybrid Cellular-Automata and Markov Model

In this study, land-cover change in the capital Conakry of Guinea was simulated using the integrated Cellular Automata and Markov model (CA-Markov) in the Geographic Information System (GIS) and Remote Sensing (RS). Historical land-cover change information was derived from 1986, 2000 and 2016 Landsat data. Using the land-cover change maps of 1986 and 2000, the land-cover change map for 2016 was simulated based on the Markov model in IDRISSI software (Clark University, Worcester, MA, USA). The simulated result was compared with the 2016 land-cover map for validation using the Relative Operating Characteristic (ROC). The ROC result showed a very strong agreement between the two maps. From this result, the land-cover change map for 2025 was simulated using CA-Markov model. The result has indicated that the proportion of the urban area was 49% in 2016, and it is expected to increase to 52% by 2025, while vegetation will decrease from 35% in 2016 to 32% in 2025. This study suggests that the rapid land-cover change has been led by both rapid population growth and extreme poverty in rural areas, which will result in migration into Conakry. The results of this study will provide bases for assessing the sustainability and the management of the urban area and for taking actions to mitigate the degradation of the urban environment.

Identification of Drought Occurrences Using Ensemble Predictions up to 20-Days in Advance

In the present study, skill of an extended range forecast system has been evaluated for identifying droughts over central India 20-days in advance. Rainfall forecasts from 44 ensemble members of the forecast system developed Indian Institute of Tropical Meteorology (IITM), Pune have been used to prepare probabilistic rainfall forecasts. It is seen that the uncertainties in the forecasts (in terms of ensemble spread) increases from day-5 to day 20. As the focus of the study is for drought predictions, forecasts in the bins 0-5 mm/5 day and 5-25 mm/5 day (no rain or less rain) were studied in detail. It is found that the modeling system has a tendency to over-forecast rainfall probabilities. With bias correction, the forecasts become more reliable. Various drought indices were computed using the mean of the forecast distribution up to 20-days in advance. Standardized precipitation index (SPI) computed using Gamma and Pearson type-III distributions are similar in the study region. It was found that these are in reasonable agreement with those from observations. Probabilistic forecasts of standardized precipitation index (SPI) were made and the relative operating characteristics (ROC) scores indicate that the forecasted SPI values are suitable for application.

Global Assessment of Atmospheric River Prediction Skill

Abstract Atmospheric rivers (ARs) are global phenomena that transport water vapor horizontally and are associated with hydrological extremes. In this study, the Atmospheric River Skill (ATRISK) algorithm is introduced, which quantifies AR prediction skill in an object-based framework using Subseasonal to Seasonal (S2S) Project global hindcast data from the European Centre for Medium-Range Weather Forecasts (ECMWF) model. The dependence of AR forecast skill is globally characterized by season, lead time, and distance between observed and forecasted ARs. Mean values of daily AR prediction skill saturate around 7–10 days, and seasonal variations are highest over the Northern Hemispheric ocean basins, where AR prediction skill increases by 15%–20% at a 7-day lead during boreal winter relative to boreal summer. AR hit and false alarm rates are explicitly considered using relative operating characteristic (ROC) curves. This analysis reveals that AR forecast utility increases at 10-day lead over the North Pacific/western U.S. region during positive El Niño–Southern Oscillation (ENSO) conditions and at 7- and 10-day leads over the North Atlantic/U.K. region during negative Arctic Oscillation (AO) conditions and decreases at a 10-day lead over the North Pacific/western U.S. region during negative Pacific–North America (PNA) teleconnection conditions. Exceptionally large increases in AR forecast utility are found over the North Pacific/western United States at a 10-day lead during El Niño + positive PNA conditions and over the North Atlantic/United Kingdom at a 7-day lead during La Niña + negative PNA conditions. These results represent the first global assessment of AR prediction skill and highlight climate variability conditions that modulate regional AR forecast skill.

Assessing the Skill of Convection-Allowing Ensemble Forecasts of Precipitation by Optimization of Spatial-Temporal Neighborhoods

The current neighborhood probability (NP) method mainly considers the spatial displacement error in high-resolution precipitation forecasts, but the problem of the forecast time exceeding or lagging behind the observed field has not been properly solved. Therefore, a temporal factor was introduced into the NP method in this paper, and precipitation forecasts were evaluated in different spatial-temporal neighborhoods based on the improved NP method and fractions skill score (FSS), combined with the relative operating characteristic (ROC) curve. The results indicated that the forecasting accuracy of the ensemble forecast was higher than the control forecast. The neighborhood ensemble probability (NEP) and probability matched mean (PMM) methods were superior to the traditional ensemble mean (EM) method in forecasting heavy rainfall, which compensated for the limitations of the heavy rainfall forecasting ability of EM. For such squall line processes, a spatial scale of 15–45 km neighborhood radius could effectively rectify the displacement error of precipitation. There was a corresponding relationship between temporal scale and rainfall intensity in convective-scale precipitation forecast, so the temporal uncertainty of different levels of precipitation could be captured by different temporal scales. The spatial and temporal scales had interdependent influences on precipitation forecast effects, which could be affected by the intrinsic spatial-temporal scale of convective-scale weather systems. The improved NP method could simultaneously reflect the spatial and temporal uncertainties of convective-scale precipitation forecasts in high-resolution models, achieving a comprehensive assessment of spatial-temporal scale and providing probabilistic forecast results that match different levels of precipitation.

Evaluation of ecosystem service based on scenario simulation of land use in Yunnan Province

Climate change and rapid urbanization are important factors restricting future land use. Situational analysis, as an important foundation for the optimization of land use, needs to focus on the impact of climate factors and socio-economic factors. In this paper, the Markov model and the DLS (Simulation of Land System Dynamics) model are combined for the first time, and the land use pattern in 2020 is simulated based on the data of land use in 2000 and 2010 as well as the climate, soil, topography and socio-economic factors of Yunnan Province. In his paper, we took Yunnan Province as the case study area, and selected 12 driving factors by logistic regression method, then the land use demands and layout of Yunnan Province in 2020 has been forecasted and simulated under business as usual (BAU) scenario and farmland protection (FP) scenario and the changes in ecosystem service value has been calculated. The result shows that: (1) after the regression analysis and ROC (Relative Operating Characteristics) test, the 12 factors selected in this paper have a strong ability to explain the land use change in Yunnan Province. (2) Under the two scenarios, the significant reduction of arable land area is a common feature of land use change in Yunnan Province in the future, and its main land use type will be construction land. However, under FP scenario, the current situation where construction land encroach on arable land will be improved. Compared with the change from 2000 to 2010, the trend of arable land, forest land, water area, construction land and unused land will be the same under the two scenarios, whereas the change trend of grassland was opposite. (3) From 2000 to 2020, the value of ecosystem services in Yunnan Province is on the rise, but the ecosystem service value under FP scenario is higher than that of the ecosystem services under BAU scenario. In general, land use in 2020 in Yunnan Province continues the pattern of 2010, but there are also significant spatial differences. Under the BAU scenario, the construction land is mainly in the south of Lijiang City and the northeastern part of Kunming. Under the FP scenario, the new construction land is concentrated near the Lashi dam in northern Yunnan Province, and the high-quality arable land in the valley will be better protected. The research results can provide reference for the optimization of land use pattern in Yunnan Province, and provide scientific basis for land use management and planning. Based on the value of ecosystem services, we should implement the policy of strict protection of arable land, both to ensure food supply and promote the healthy development of ecological environment.

On the Directional Accuracy of United States Housing Starts Forecasts: Evidence from Survey Data

I use data from both the Survey of Professional Forecasters and the Livingston Survey to study the directional accuracy of United States housing starts forecasts. Using elements of relative operating characteristic (ROC) analysis, I find that forecasts contain information with respect to subsequent changes in housing starts. Estimates for both surveys are significant at all forecast horizons and robust across time and across forecasters. Implications for the usage of housing starts forecasts from survey data are discussed.

Spatial modelling of gully erosion in Mazandaran Province, northern Iran

Gully erosion is one of the most severe environmental problems in large areas of Iran. The spatial distribution of gully erosion and its susceptibility zonation was studied using different bivariate statistical models, such as frequency ratio (FR), weights of evidence (WofE), and index of entropy (IofE). For this purpose, 109 gully erosion locations were identified and divided into training (70%) and validating (30%) datasets. Effective factors, including elevation, slope aspect, slope degree, slope-length (LS), topographical wetness index (TWI), plan curvature, profile curvature, land use, lithology, distance from river, drainage density, and distance from road were selected to develop maps of gully erosion susceptibility. The spatial relationship between gully erosion and each effective factor was calculated by the mentioned models. The relative operating characteristic (ROC) curve was implemented for evaluating the accuracy of the applied predictive models. Results indicated that frequency ratio model had better performance (80.4%) than the weight of evidence (79.5%) and index of entropy (79%) models. The produced gully erosion susceptibility maps can be helpful to make decisions for soil and water planning and management and finally sustainable development in the Valasht watershed.

Rainfall prediction of Cimanuk watershed regions with canonical correlation analysis (CCA)

Rainfall prediction in Indonesia is very influential on various development sectors, such as agriculture, fisheries, water resources, industry, and other sectors. The inaccurate predictions can lead to negative effects. Cimanuk watershed is one of the main pillar of water resources in West Java. This watersheds divided into three parts, which is a headwater of Cimanuk sub-watershed, Middle of Cimanuk sub-watershed and downstream of Cimanuk sub- watershed. The flow of this watershed will flow through the Jatigede reservoir and will supply water to the north-coast area in the next few years. So, the reliable model of rainfall prediction is very needed in this watershed. Rainfall prediction conducted with Canonical Correlation Analysis (CCA) method using Climate Predictability Tool (CPT) software. The prediction is every 3months on 2016 (after January) based on Climate Hazards group Infrared Precipitation with Stations (CHIRPS) data over West Java. Predictors used in CPT were the monthly data index of Nino3.4, Dipole Mode (DMI), and Monsoon Index (AUSMI-ISMI-WNPMI-WYMI) with initial condition January. The initial condition is chosen by the last data update. While, the predictant were monthly rainfall data CHIRPS region of West Java. The results of prediction rainfall showed by skill map from Pearson Correlation. High correlation of skill map are on MAM (Mar-Apr-May), AMJ (Apr-May-Jun), and JJA (Jun-Jul-Aug) which means the model is reliable to forecast rainfall distribution over Cimanuk watersheds region (over West Java) on those seasons. CCA score over those season prediction mostly over 0.7. The accuracy of the model CPT also indicated by the Relative Operating Characteristic (ROC) curve of the results of Pearson correlation 3 representative point of sub-watershed (Sumedang, Majalengka, and Cirebon), were mostly located in the top line of non-skill, and evidenced by the same of rainfall patterns between observation and forecast. So, the model of CPT with CCA method is reliable to use.

Modelling and mapping of soil damage caused by harvesting in Caspian forests (Iran) using CART and RF data mining techniques

Controlling the soil damage caused by forest harvesting has a key role in forest management due to its effect on forest dynamics and productivity, mainly through modifying the physical, mechanical, and hydrological context of soil. This study was conducted to evaluate the soil damage susceptibility in one of the Caspian forests, Iran. For this purpose, two data mining techniques including classification and regression tree (CART) and random forest (RF) were applied. A total of 224 soil damage locations were identified primarily from field surveys. Then, 10 conditioning variables were produced in GIS. For model performance, the outputs of the analyses were compared with the field-verified soil damage locations. Our results show that slope degree, soil type, and slope aspect had the highest weight on soil damage, in the order of their appurtenance. Additionally, according to the relative operating characteristics curve, RF is a more suitable prediction model for soil damage zoning compared to CART. In summary, the findings of this study suggest that soil damage susceptibility mapping is an effective technique for Caspian forests, Iran.

Comparison of Next-Day Probabilistic Severe Weather Forecasts from Coarse- and Fine-Resolution CAMs and a Convection-Allowing Ensemble

Abstract Given increasing computing power, an important question is whether additional computational resources would be better spent reducing the horizontal grid spacing of a convection-allowing model (CAM) or adding members to form CAM ensembles. The present study investigates this question as it applies to CAM-derived next-day probabilistic severe weather forecasts created by using forecast updraft helicity as a severe weather proxy for 63 days of the 2010 and 2011 NOAA Hazardous Weather Testbed Spring Forecasting Experiments. Forecasts derived from three sets of Weather Research and Forecasting Model configurations are tested: a 1-km deterministic model, a 4-km deterministic model, and an 11-member, 4-km ensemble. Forecast quality is evaluated using relative operating characteristic (ROC) curves, attributes diagrams, and performance diagrams, and forecasts from five representative cases are analyzed to investigate their relative quality and value in a variety of situations. While no statistically significant differences exist between the 4- and 1-km deterministic forecasts in terms of area under ROC curves, the 4-km ensemble forecasts offer weakly significant improvements over the 4-km deterministic forecasts over the entire 63-day dataset. Further, the 4-km ensemble forecasts generally provide greater forecast quality relative to either of the deterministic forecasts on an individual day. Collectively, these results suggest that, for purposes of improving next-day CAM-derived probabilistic severe weather forecasts, additional computing resources may be better spent on adding members to form CAM ensembles than on reducing the horizontal grid spacing of a deterministic model below 4 km.

Good employees through good jobs

PurposeThe purpose of this paper is to identify different job types in the Belgian electricity sector and their relations with employee outcomes such as work engagement and innovative work behaviour (IWB).Design/methodology/approachThis paper uses a combination of latent profile analysis and relative operating characteristics (ROC) analysis.FindingsDepending on the job resources and demands, five different job types are identified corresponding largely to the Karasek and Theorell (1990) job types. Their relation with the outcomes is not parallel with low-strain jobs performing best for work engagement, and active jobs for IWB.Research limitations/implicationsThe combination of methods used in this study increases significantly the ease of communication of the findings, yet an external benchmark for the ROC analysis would be preferable.Practical implicationsTo foster engagement and IWB with employees one should focus on the job content and only increase demands if they are combined with sufficient resources.Originality/valueThis research is the first in its kind that relates latent job types with different employee outcomes using a combination of latent profile and ROC analysis.

Validation and ranking of seven staging systems of hepatocellular carcinoma.

The aim of the present study was to evaluate the ability of seven staging systems to predict 3- and 6-month and cumulative survival rates of patients with advanced hepatitis B virus (HBV)-associated hepatocellular carcinoma (HCC). Data were collected from 220 patients with HBV-associated HCC who did not receive any standard anticancer treatment. Participants were patients at The Third Affiliated Hospital of Sun Yat-sen University from September 2008 to June 2010. The participants were classified according to the Chinese University Prognostic Index (CUPI), the Cancer of the Liver Italian Program (CLIP), Japan Integrated Staging (JIS), China Integrated Score (CIS) systems, Barcelona Clinic Liver Cancer (BCLC), Okuda and tumor-node-metastasis (TNM) staging systems at the time of diagnosis and during patient follow-up. The sensitivity and specificity of the predictive value of each staging system for 3- and 6-month mortality were analyzed by relative operating characteristic (ROC) curve analysis with a non-parametric test being used to compare the area under curve (AUC) of the ROC curves. In addition, log-rank tests and Kaplan-Meier estimator survival curves were applied to compare the overall survival rates of the patients with HCC defined as advanced using the various staging systems, and the Akaike information criterion (AIC) and likelihood ratio tests (LRTs) were used to evaluate the predictive value for overall survival in patients with advanced HCC. Using univariate and multivariate Cox's model analyses, the factors predictive of survival were also identified. A total of 220 patients with HBV-associated HCC were analyzed. Independent prognostic factors identified by multivariate analyses included tumor size, α-fetoprotein levels, blood urea nitrogen levels, the presence or absence of portal vein thrombus, Child-Pugh score and neutrophil count. When predicting 3-month survival, the AUCs of CLIP, CIS, CUPI, Okuda, TNM, JIS and BCLC were 0.806, 0.772, 0.751, 0.731, 0.643, 0.754 and 0.622, respectively. When predicting 6-month survival, the AUCs of CLIP, CIS, CUPI, Okuda, TNM, JIS and BCLC were 0.828, 0.729, 0.717, 0.692, 0.664, 0.746 and 0.575, respectively. For 3-month mortality, the prognostic value of CLIP ranked highest, followed by CIS; for 6-month mortality, the prognostic value of CLIP also ranked highest, followed by JIS. No significant difference between the AUCs of CLIP and CIS (P>0.05) in their predictive value for 3-month mortality was observed. The AUC of CLIP was significantly higher compared with that of the other staging systems (P<0.05) for predicting 6-month mortality. The χ2 values from the LRTs of CLIP, CIS, CUPI, Okuda, TNM, JIS and BCLC were 75.6, 48.4, 46.7, 36.0, 21.0, 46.8 and 7.24, respectively. The AIC values of CLIP, CIS, CUPI, Okuda, TNM, JIS and BCLC were 1601.5, 1632.3, 1629.9, 1641.1, 1654.8, 1627.4 and 1671.1, respectively. CLIP exhibited the highest χ2 value and lowest AIC value, indicating that CLIP has the highest predictive value of cumulative survival rate. In the selected patients of the present study, CLIP was the staging system best able to predict 3- and 6-month and overall survival rates. CIS ranked second in predicting 3-month mortality.

On the directional accuracy of inflation forecasts: evidence from South African survey data

ABSTRACTWe study the information content of South African inflation survey data by determining the directional accuracy of both short-term and long-term forecasts. We use relative operating characteristic (ROC) curves, which have been applied in a variety of fields including weather forecasting and radiology, to ascertain the directional accuracy of the forecasts. A ROC curve summarizes the directional accuracy of forecasts by comparing the rate of true signals (sensitivity) with the rate of false signals (one minus specifity). A ROC curve goes beyond market-timing tests widely studied in earlier research as this comparison is carried out for many alternative values of a decision criterion that discriminates between signals (of a rising inflation rate) and nonsignals (of an unchanged or a falling inflation rate). We find consistent evidence that forecasts contain information with respect to the subsequent direction of change of the inflation rate.

Landslide susceptibility assessment using the certainty factor and analytic hierarchy process

A new approach combining the certainty factor (CF) and analytic hierarchy process (AHP) methods was proposed to assess landslide susceptibility in the Ziyang district, which is situated in the Qin-Ba Mountain region, China. Landslide inventory data were collected based on field investigations and remote sensing interpretations. A total of 791 landslides were identified. A total of 633 landslides were randomly selected from this data set as the training set, and the remaining landslides were used for validation as the test set. Nine factors, including the slope angle, slope aspect, slope curvature, lithology, distance to faults, distance to streams, precipitation, road network intensity degree and land use were chosen as the landslide causal factors for further susceptibility assessment. The weight of each factor and its subclass were calculated by AHP and CF methods. Landslide susceptibility was compared between the bivariate statistical method and the proposed CF-AHP method. The results indicate that the distance to streams, distance to faults and lithology are the most dominant causal factors associated with landslides. The susceptibility zonation was categorized into five classes of landslide susceptibility, i.e., very high, high, moderate, low and very low level. Lastly, the relative operating characteristics (ROC) curve was used to validate the accuracy of the new approach, and the result showed a satisfactory prediction rate of 78.3%, compared to 69.2% obtained with the landslide susceptibility index method. The results indicate that the CF-AHP combined method is more appropriate for assessing the landslide susceptibility in this area.

Real-time Alarm Monitoring System for Detecting Driver Fatigue in Wireless Areas

The purpose of this paper was to develop a real-time alarm monitoring system that can detect the fatigue driving state through wireless communication. The drivers’ electroencephalogram (EEG) signals were recorded from occipital electrodes. Seven EEG rhythms with different frequency bands as gamma, hbeta, beta, sigma, alpha, theta and delta waves were extracted. They were simultaneously assessed using relative operating characteristic (ROC) curves and grey relational analysis to select one as the fatigue feature. The research results showed that the performance of theta wave was the best one. Therefore, theta wave was used as fatigue feature in the following alarm device. The real-time alarm monitoring system based on the result has been developed, once the threshold was settled by using the data of the first ten minutes driving period. The developed system can detect driver fatigue and give alarm to indicate the onset of fatigue automatically.

Modeling Determinants of Urban Growth in Conakry, Guinea: A Spatial Logistic Approach

The main objective of the present study was to integrate a logistic regression model (LRM), a geographic information system (GIS) and remote sensing (RS) techniques to analyze and quantify urban growth patterns and investigate the relationship between urban growth and various driving forces. Landsat images from 1986, 2000, and 2016 derived from the TM, ETM+, and OLI sensors respectively were used to simulate an urban growth probability map for Conakry. To better explain the effects of the drivers on the urban growth processes in the study area, variables for two groups of drivers were considered: socioeconomic proximity and physical topography. The results of the LRM using IDRISI Selva indicated that the variables elevation (β7 = 1.76) and distance to major roads (β4 = 0.67) resulted in models with the best fit and the highest regression coefficients. These results indicate a high probability of urban growth in areas with high elevation and near major roads. The validation of the model was conducted using the relative operating characteristic (ROC) method; which result exhibited high accuracy of 0.89 between the simulated urban growth probability map and the actual one. A land use/land cover (LULC) change analysis showed that the urban area had undergone continuous growth over the study period resulting in an extent of 143.5 km2 for the urban area class in 2016.

Development of a GIS-based catastrophe theory model (modified DRASTIC model) for groundwater vulnerability assessment

This study developed a new paradigm for groundwater vulnerability assessment by modifying the standard DRASTIC index (DI) model based on catastrophe theory. The developed paradigm was called the catastrophe theory-based DI (CDI) model. The proposed model was applied to assess groundwater vulnerability to pollution index (GVPI) in Perak Province, Malaysia. The area vulnerability index was modeled by considering the DRASTIC multiple vulnerability causative factors (VCFs) obtained from different data sources. The weights and ranking of the VCFs were computed by using the inner fuzzy membership mechanism of the CDI model. The estimated vulnerability index values of the CDI model were processed in a geographic information system (GIS) environment to produce a catastrophe theory–DRASTIC groundwater vulnerability to pollution index (CDGVPI) map, which demarcated the area into five vulnerability zones. The produced CDGVPI map was validated by applying the water quality status–vulnerability zone relationship (WVR) approach and the relative operating characteristic (ROC) curve method. The performance of the developed CDI model was compared with that of the standard DI model. The validation results of the WVR approach exhibits 89.29% prediction accuracy for the CDI model compared with 75% for the DI model. Meanwhile, the ROC validation results for the CDI and DI models are 88.8% and 78%, respectively. The GIS-based CDI model demonstrated better performance than the DI model. The GVPI maps produced in this study can be used for precise decision making process in environmental planning and groundwater management.

Application of Bivariate and Multivariate Statistical Techniques in Landslide Susceptibility Modeling in Chittagong City Corporation, Bangladesh

The communities living on the dangerous hillslopes in Chittagong City Corporation (CCC) in Bangladesh recurrently experience landslide hazards during the monsoon season. The frequency and intensity of landslides are increasing over time because of heavy rainfall occurring over a few days. Furthermore, rapid urbanization through hill-cutting is another factor, which is believed to have a significant impact on the occurrence of landslides. This study aims to develop landslide susceptibility maps (LSMs) through the use of Dempster-Shafer weights of evidence (WoE) and the multiple regression (MR) method. Three different combinations with principal component analysis (PCA) and fuzzy membership techniques were used and tested. Twelve factor maps (i.e., slope, hill-cutting, geology, geomorphology, NDVI, soil moisture, precipitation and distance from existing buildings, stream, road and drainage network, and faults-lineaments) were prepared based on their association with historical landslide events. A landslide inventory map was prepared through field surveys for model simulation and validation purposes. The performance of the predicted LSMs was validated using the area under the relative operating characteristic (ROC) curve method. The overall success rates were 87.3%, 90.9%, 91.3%, and 93.9%, respectively for the WoE, MR with all the layers, MR with PCA layers, and MR with fuzzy probability layers.

Improving the capability of an integrated CA-Markov model to simulate spatio-temporal urban growth trends using an Analytical Hierarchy Process and Frequency Ratio

The creation of an accurate simulation of future urban growth is considered one of the most important challenges in urban studies that involve spatial modeling. The purpose of this study is to improve the simulation capability of an integrated CA-Markov Chain (CA-MC) model using CA-MC based on the Analytical Hierarchy Process (AHP) and CA-MC based on Frequency Ratio (FR), both applied in Seremban, Malaysia, as well as to compare the performance and accuracy between the traditional and hybrid models. Various physical, socio-economic, utilities, and environmental criteria were used as predictors, including elevation, slope, soil texture, population density, distance to commercial area, distance to educational area, distance to residential area, distance to industrial area, distance to roads, distance to highway, distance to railway, distance to power line, distance to stream, and land cover. For calibration, three models were applied to simulate urban growth trends in 2010; the actual data of 2010 were used for model validation utilizing the Relative Operating Characteristic (ROC) and Kappa coefficient methods Consequently, future urban growth maps of 2020 and 2030 were created. The validation findings confirm that the integration of the CA-MC model with the FR model and employing the significant driving force of urban growth in the simulation process have resulted in the improved simulation capability of the CA-MC model. This study has provided a novel approach for improving the CA-MC model based on FR, which will provide powerful support to planners and decision-makers in the development of future sustainable urban planning.

Landslide susceptibility mapping at Ovacık-Karabük (Turkey) using different artificial neural network models: comparison of training algorithms

This study aims to investigate the performances of different training algorithms used for an artificial neural network (ANN) method to produce landslide susceptibility maps. For this purpose, Ovacik region (southeast of Karabuk Province), located in the Western Black Sea Region (Turkey), was selected as the study area. A total of 196 landslides were mapped, and a landslide database was prepared. Topographical elevation, slope angle, aspect, wetness index, lithology, and vegetation index parameters were taken into account for the landslide susceptibility analyses. Two different ANN structures, which were composed of single and double hidden layers, were applied to compare the effects of the ANN. Four different training algorithms, namely batch back-propagation, quick propagation, conjugate gradient descent (CGD), and Levenberg–Marquardt, were used for the training stage of the ANN models. Thus, eight different landslide susceptibility maps were produced for the study area using different ANN structures and algorithms. In order to assess the effects and spatial performances of the considered training algorithms on the ANN models, the relative operating characteristics (ROC) and relation value (rij) approaches were used. The susceptibility map produced by CGD1 has the highest AUC (0.817) and rij values (0.972). Comparison of the susceptibility maps indicated that CGD training algorithm is the slowest one among the other algorithms, but this algorithm showed the highest performance on the results.