PMT: New analytical framework for automated evaluation of geo-environmental modelling approaches

Rahmati, Omid; Kornejady, Aiding; Samadi, Mehdi; Deo, Ravinesh C.; Conoscenti, Christian; Lombardo, Luigi; Dayal, Kavina; Taghizadeh‐Mehrjardi, Ruhollah; Pourghasemi, Hamid Reza; Kumar, Sandeep; Bui, Dieu Tien

doi:10.1016/j.scitotenv.2019.02.017

Cited by 90 publications

(49 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The FPR (1-specificity) indicate the probability of incorrect predictions of non-event location as an event. TSS also measure the ability of a predicted value to discriminate between the events and non-events, using all of the elements in the confusion matrix 42 . The CCI considers TN and FN for true-and false-negative predicted events, and TP and FP for true-and false-positive, respectively.…”

Section: Construction Of Flood Forest Fire and Landslide Conditionimentioning

confidence: 99%

Assessing and mapping multi-hazard risk susceptibility using a machine learning technique

Pourghasemi

Kariminejad

Amiri

et al. 2020

Sci Rep

Self Cite

137

View full text Add to dashboard Cite

the aim of the current study was to suggest a multi-hazard probability assessment in fars province, Shiraz City, and its four strategic watersheds. At first, we construct maps depicting the most effective factors on floods (12 factors), forest fires (10 factors), and landslides (10 factors), and used the Boruta algorithm to prioritize the impact of each respective factor on the occurrence of each hazard. Subsequently, flood, landslides, and forest fire susceptibility maps prepared using a Random Forest (RF) model in the R statistical software. Results indicate that 42.83% of the study area are not susceptible to any hazards, while 2.67% of the area is at risk of all three hazards. The results of the multi-hazard map in Shiraz City indicate that 25% of Shiraz city is very susceptible to flooding, while 16% is very susceptible to landslide occurrences. for four strategic watersheds, it is notable that in the Dorodzan Watershed, landslides and floods are the most important hazards; whereas, flood occurrences cover the largest area of the Maharlou Watershed. In contrast, the Tashk-Bakhtegan Watershed is so sensible to floods and landslides, respectively. Finally, in the Ghareaghaj Watershed, forest fire ranks as the strongest hazard, followed by floods. The validation results indicate an AUC of 0.834, 0.939, and 0.943 for the flood, landslide, and forest fire susceptibility maps, respectively. Also, other accuracy measures including, specificity, sensitivity, TSS, CCI, and Gini coefficient confirmed results of the AUC values. These results allow us to forecast the spatial behavior of such multi-hazard events, and researchers and stakeholders alike can apply them to evaluate hazards under various mitigation scenarios.The Sendai Framework, with its comprehensive vision, recommends more efforts to decrease disaster risk and increase sustainable development. Especially communities who are increasingly susceptible to natural hazards should adhere to these guidelines and plan accordingly. In this regard, the multi-hazard approach is often used in risk reduction projects and studies addressing risks associated with human activities or climate change on a regional and local scale 1 . It is obvious that introducing a universal set of multi-hazard assessment techniques is of fundamental importance for reducing disaster risk, and constitutes a valuable asset to share with other stakeholders, including the private sectors, local governments, and other stakeholders.The use of the term multi-hazard in the current research is related to the objective of risk reduction among natural hazards, including flood, landslides, and forest fires, in a specified spatial distribution in this study 2,3 . Recently, susceptibility modeling approaches related to single processes have advanced considerably for river floods 4 and landslides 5-7 . However, there is still neither a common terminology nor a uniform conceptual approach for analyzing multiple hazards in conjunction. This is not unexpected because multi-hazard analyses are not the sum of ...

show abstract

Section: Construction Of Flood Forest Fire and Landslide Conditionimentioning

confidence: 99%

Assessing and mapping multi-hazard risk susceptibility using a machine learning technique

Pourghasemi

Kariminejad

Amiri

et al. 2020

Sci Rep

Self Cite

137

View full text Add to dashboard Cite

show abstract

“…The final suitability map derived from WLC is subjected to a performance assessment test, which is carried out based on the ROC fed by locations that have already flooded in recent years. The ROC as a holdout-independent measure can decisively pinpoint the performance of any spatial model and is a crucial tool for validating the results of spatial maps [36]. It plots the 1-specificity (i.e., false positive; incorrectly predicting an unsuitable location as suitable) on the x-axis against the sensitivity (i.e., true positive; correctly predicting a suitable location as observed in reality) on the y-axis [37].…”

Section: Designing the Site Selection Software (Sss)mentioning

confidence: 99%

“…The AUC value varies between 0 and 1; a higher value implies a higher prediction performance, whereas a value near 0.5 indicates that the prediction is no better than a random prediction [73]. According to Rahmati et al [36], the strength of agreement given the AUC magnitude is for 0-0.2 slight, 0.2-0.4 fair, 0.4-0.6 moderate, 0.6-0.8 substantial, and 0.8-1.0 almost perfect. The SSS can plot the ROC curve based on training and validation data sets to determine the goodness-of-fit and predictive performance of site selection methods.…”

Section: Accuracy Assessmentmentioning

confidence: 99%

GIS-Based Site Selection for Check Dams in Watersheds: Considering Geomorphometric and Topo-Hydrological Factors

et al. 2019

Self Cite

View full text Add to dashboard Cite

Check dams are widely used watershed management measures for reducing flood peak discharge and sediment transport, and increasing lag time and groundwater recharge throughout the world. However, identifying the best suitable sites for check dams within the stream networks of various watersheds remains challenging. This study aimed to develop an open-source software with user-friendly interface for screening the stream network possibilities and identifying and guiding the selection of suitable sites for check dams within watersheds. In this developed site selection software (SSS), multi-criteria decision analysis (MCDA) was integrated into geographic information systems (GIS), which allowed for numerous spatial data of the multiple criteria to be relatively simply and visually processed. Different geomorphometric and topo-hydrological factors were considered and accounted for to enhance the SSS identification of the best locations for check dams. The factors included topographic wetness index (TWI), terrain ruggedness index (TRI), topographic position index (TPI), sediment transport index (STI), stream power index (SPI), slope, drainage density (DD), and stream order (SO). The site identification performance of the SSS was assessed using the receiver operating characteristic (ROC) curve method, with results for the case study example of the Poldokhtar watershed in Iran showing excellent performance and identifying 327 potential sites for efficient check dam construction in this watershed. The SSS tool is not site-specific but is rather general, adaptive, and comprehensive, such that it can and should be further applied and tested across different watersheds and parts of the world.

show abstract

“…The closer the AUC-ROC value is to 1, the better the model performance. All evaluation criteria were calculated using the performance measure tool (PMT) extension [77], which allows learning capability (also termed goodness-of-fit) and predictive performance to be determined based on the training and validation datasets, respectively.…”

Section: Accuracy Assessmentmentioning

confidence: 99%

An Automated Python Language-Based Tool for Creating Absence Samples in Groundwater Potential Mapping

et al. 2019

Self Cite

View full text Add to dashboard Cite

Although sampling strategy plays an important role in groundwater potential mapping and significantly influences model accuracy, researchers often apply a simple random sampling method to determine absence (non-occurrence) samples. In this study, an automated, user-friendly geographic information system (GIS)-based tool, selection of absence samples (SAS), was developed using the Python programming language. The SAS tool takes into account different geospatial concepts, including nearest neighbor (NN) and hotspot analyses. In a case study, it was successfully applied to the Bojnourd watershed, Iran, together with two machine learning models (random forest (RF) and multivariate adaptive regression splines (MARS)) with GIS and remotely sensed data, to model groundwater potential. Different evaluation criteria (area under the receiver operating characteristic curve (AUC-ROC), true skill statistic (TSS), efficiency (E), false positive rate (FPR), true positive rate (TPR), true negative rate (TNR), and false negative rate (FNR)) were used to scrutinize model performance. Two absence sample types were produced, based on a simple random method and the SAS tool, and used in the models. The results demonstrated that both RF (AUC-ROC = 0.913, TSS = 0.72, E = 0.926) and MARS (AUC-ROC = 0.889, TSS = 0.705, E = 0.90) performed better when using absence samples generated by the SAS tool, indicating that this tool is capable of producing trustworthy absence samples to improve groundwater potential models.

show abstract

PMT: New analytical framework for automated evaluation of geo-environmental modelling approaches

Cited by 90 publications

References 51 publications

Assessing and mapping multi-hazard risk susceptibility using a machine learning technique

Assessing and mapping multi-hazard risk susceptibility using a machine learning technique

GIS-Based Site Selection for Check Dams in Watersheds: Considering Geomorphometric and Topo-Hydrological Factors

An Automated Python Language-Based Tool for Creating Absence Samples in Groundwater Potential Mapping

Contact Info

Product

Resources

About