Urban Flood Risk Assessment: Integration of Decision Making and Machine Learning

Urban flood risk mapping is an important tool for the mitigation of flooding in view of human activities and climate change. Many developing countries, however, lack sufficiently detailed data to produce reliable risk maps with existing methods. Thus, improved methods are needed that can improve urban flood risk management in regions with scarce hydrological data. Given this, we estimated the flood risk map for Rasht City (Iran), applying a composition of decision-making and machine learning methods. Flood hazard maps were produced applying six state-of-the-art machine learning algorithms such as: classification and regression trees (CART), random forest (RF), boosted regression trees (BRT), multivariate adaptive regression splines (MARS), multivariate discriminant analysis (MDA), and support vector machine (SVM). Flood conditioning parameters applied in modeling were elevation, slope angle, aspect, rainfall, distance to river (DTR), distance to streets (DTS), soil hydrological group (SHG), curve number (CN), distance to urban drainage (DTUD), urban drainage density (UDD), and land use. In total, 93 flood location points were collected from the regional water company of Gilan province combined with field surveys. We used the Analytic Hierarchy Process (AHP) decision-making tool for creating an urban flood vulnerability map, which is according to population density (PD), dwelling quality (DQ), household income (HI), distance to cultural heritage (DTCH), distance to medical centers and hospitals (DTMCH), and land use. Then, the urban flood risk map was derived according to flood vulnerability and flood hazard maps. Evaluation of models was performed using receiver-operator characteristic curve (ROC), accuracy, probability of detection (POD), false alarm ratio (FAR), and precision. The results indicated that the CART model is most accurate model (AUC = 0.947, accuracy = 0.892, POD = 0.867, FAR = 0.071, and precision = 0.929). The results also demonstrated that DTR, UDD, and DTUD played important roles in flood hazard modeling; whereas, the population density was the most significant parameter in vulnerability mapping. These findings indicated that machine learning methods can improve urban flood risk management significantly in regions with limited hydrological data.

Agricultural Flood Vulnerability Assessment and Risk Quantification in Iowa

Agricultural lands are often impacted by flooding, which results in economic losses and causes food insecurity across the world. Due to the world’s growing population, land-use alteration is frequently practiced to meet global demand. However, land-use changes combined with climate change have resulted in extreme hydrological changes (i.e., flooding and drought) in many areas. The state of Iowa has experienced several flooding events over the last couple of decades (e.g., 1993, 2008, 2014, 2016, 2019). Also, agribusiness is conducted across 85 percent of the state. In this research, we present a comprehensive assessment for agricultural flood risk in the state of Iowa utilizing most up-to-date flood inundation maps and crop layer raster datasets. The study analyzes the seasonal variation of the statewide agricultural flood risk by focusing on corn, soybean, and alfalfa crops. It also investigates the crop frequency layers and corn suitability rating datasets to reveal regions with lower or higher productivity ratings. Additionally, a terrain-based flood model is used to analyze performance against the FEMA maps. The research discusses the potential mitigation activities for the most vulnerable watersheds in the state. The analysis shows that nearly a half-million acres of cornfields and soybean fields are located in the 2-year flood zone. We also found that terrain-based flood maps are a reliable alternative for agricultural flood risk assessment based on their dynamic structure, rapid update capability, and performance compared to FEMA maps.

Mitigasi Dan Adaptasi Bencana Banjir di Kecamatan Pallangga Kabupaten Gowa

Bencana banjir yang terjadi secara terus-menerus dapat menyebabkan berbagai kondisi yang apabila terjadi dapat menimbulkan berbagai kerentanan yang memerlukan pemikiran yang lebih dalam untuk mengantisipasi bencana banjir. Mitigasi dan Adaptasi Bencana Banjir di Kecamatan Pallangga Kabupaten Gowa bertujuan untuk memetakan tingkat kerentanan bencana banjir dan upaya mitigasi dan adaptasi yang tepat berdasarakan tingkat kerentanan bencana banjir di Kecamatan Pallangga. Sejalan dengan tujuan penelitian ini maka dalam penelitian ini menggunakan metode penelitian deskriptif kualitatif, untuk menentukan tingkat kerentanan dengan analisis spasial overlay dan skoring parameter penentu kerentanan banjir. Hasil dari penelitian ini adalah Kecamatan Pallangga diklasifikasi dalam tiga tingkat kerentanan bencana banjir meliputi rentan tinggi, rentan sedang dan rentan rendah serta arahan mitigasi dan adaptasi bencana banjir berdasarkan tingkat kerentanan. The floods disasters that occur continuously can cause various conditions which can cause a variety of vulnerabilities that require deeper thought to anticipate. Flood mitigation and adaptation in Pallangga District of Gowa aims to map the level of flood vulnerability and provide mitigation and adaptation directions based on the level of flood vulnerability in Pallangga. This research uses descriptive qualitative research methods, to determine the level of vulnerability with spatial overlay analysis and weight scoring of parameters determining flood vulnerability. The results of this research are in Pallangga classified into three levels of vulnerability to flood disasters including high vulnerability, medium vulnerability and low vulnerability with directives flood mitigation and adaptation based on the level of vulnerability.

Possibilistic response surfaces: incorporating fuzzy thresholds into bottom-up flood vulnerability analysis

Several alternatives have been proposed to shift the paradigms of water management under uncertainty from predictive to decision-centric. An often-mentioned tool is the response surface mapping system performance with a large sample of future hydroclimatic conditions through a stress test. Dividing this exposure space between acceptable and unacceptable states requires a criterion of acceptable performance defined by a threshold. In practice, however, stakeholders and decision-makers may be confronted with ambiguous objectives for which the acceptability threshold is not clearly defined (crisp). To accommodate such situations, this paper integrates fuzzy thresholds to the response surface tool. Such integration is not straightforward when response surfaces also have their own irreducible uncertainty from the limited number of descriptors and the stochasticity of hydroclimatic conditions. Incorporating fuzzy thresholds, therefore, requires articulating categories of imperfect knowledge that are different in nature, i.e., the irreducible uncertainty of the response itself relative to the variables that describe change and the ambiguity of the acceptability threshold. We, thus, propose possibilistic surfaces to assess flood vulnerability with fuzzy acceptability thresholds. An adaptation of the logistic regression for fuzzy set theory combines the probability of an acceptable outcome and the ambiguity of the acceptability criterion within a single possibility measure. We use the flood-prone reservoir system of the Upper Saint François River basin in Canada as a case study to illustrate the proposed approach. Results show how a fuzzy threshold can be quantitatively integrated when generating a response surface and how ignoring it might lead to different decisions. This study suggests that further conceptual developments could link the reliance on acceptability thresholds in bottom-up assessment frameworks with the current uses of fuzzy set theory.

Multivariate Analysis and Machine Learning Approach for Mapping the Variability and Vulnerability of Urban Flooding: The Case of Tangier City, Morocco

Urban flooding is a complex natural hazard, driven by the interaction between several parameters related to urban development in a context of climate change, which makes it highly variable in space and time and challenging to predict. In this study, we apply a multivariate analysis method (PCA) and four machine learning algorithms to investigate and map the variability and vulnerability of urban floods in the city of Tangier, northern Morocco. Thirteen parameters that could potentially affect urban flooding were selected and divided into two categories: geo-environmental parameters and socio-economic parameters. PCA processing allowed identifying and classifying six principal components (PCs), totaling 73% of the initial information. The scores of the parameters on the PCs and the spatial distribution of the PCs allow to highlight the interconnection between the topographic properties and urban characteristics (population density and building density) as the main source of variability of flooding, followed by the relationship between the drainage (drainage density and distance to channels) and urban properties. All four machine learning algorithms show excellent performance in predicting urban flood vulnerability (ROC curve > 0.9). The Classifications and Regression Tree and Support Vector Machine models show the best prediction performance (ACC = 91.6%). Urban flood vulnerability maps highlight, on the one hand, low lands with a high drainage density and recent buildings, and on the other, higher, steep-sloping areas with old buildings and a high population density, as areas of high to very-high vulnerability.

Vulnerability and Adaptation to Flood Hazards in Rural Settlements of Limpopo Province, South Africa

Climate change has increased the frequency of extreme weather events such as heavy rainfall leading to floods in several regions. In Africa, rural communities are more vulnerable to flooding, particularly those that dwell in low altitude areas or near rivers and those regions affected by tropical storms. This study examined flood vulnerability in three rural villages in South Africa’s northern Limpopo Province and how communities are building resilience and coping with the hazard. These villages lie at the foot of the north-eastern escarpment, and are often exposed to frequent rainfall enhanced by orographic factors. Although extreme rainfall events are rare in the study area, we analyzed daily rainfall and showed how heavy rainfall of short duration can lead to flooding using case studies. Historical floods were also mapped using remote sensing via the topographical approach and two types of flooding were identified, i.e., those due to extreme rainfall and those due to poor drainage or blocked drainage channels. A field survey was also conducted using questionnaires administered to samples of affected households to identify flood vulnerability indicators and adaptation strategies. Key informant interviews were held with disaster management authorities to provide additional information on flood indicators. Subsequently, a flood vulnerability index was computed to measure the extent of flood vulnerability of the selected communities and it was found that all three villages have a ‘vulnerability to floods’ level, considered a medium level vulnerability. The study also details temporary and long-term adaptation strategies/actions employed by respondents and interventions by local authorities to mitigate the impacts of flooding. Adaptation strategies range from digging furrows to divert water and temporary relocations, to constructing a raised patio around the house. Key recommendations include the need for public awareness; implementation of a raft of improvements and a sustainable infrastructure maintenance regime; integration of modern mitigations with local indigenous knowledge; and development of programs to ensure resilience through incorporation of Integrated Development Planning.