Abstract In recent decades, a striking number of countries have suffered from consecutive disasters: events whose impacts overlap both spatially and temporally, while recovery is still under way. The risk disasters will increase due to growing exposure, the interconnectedness human society, increased frequency intensity nontectonic hazard. This paper provides an overview different types disasters, their causes, impacts. can be distinctly occurring in isolation (both temporally) other noting...

Automated classification of building damage in remote sensing images enables the rapid and spatially extensive assessment impact natural hazards, thus speeding up emergency response efforts. Convolutional neural networks (CNNs) can reach good performance on such a task experimental settings. How CNNs perform when applied under operational conditions, with unseen data time constraints, is not well studied. This study focuses applicability CNN-based model scenarios. We performed experiments 13...

People possess a creative set of strategies based on their local knowledge (LK) that allow them to stay in flood-prone areas. Stakeholders involved with level flood risk management (FRM) often overlook and underutilise this LK. There is thus an increasing need for its identification, documentation assessment. Based qualitative research, paper critically explores the notion LK Malawi. Data was collected through 15 focus group discussions, 36 interviews field observation, analysed using...

Food insecurity is a growing concern due to man-made conflicts, climate change, and economic downturns. Forecasting the state of food essential be able trigger early actions, for example, by humanitarian actors. To measure actual insecurity, expert consensus-based approaches surveys are currently used. Both require substantial manpower, time, budget. This paper introduces an extreme gradient-boosting machine learning model forecast monthly transitions in security Ethiopia, at spatial...

The presence of biases has been demonstrated in a wide range machine learning applications; however, it is not yet widespread the case geospatial datasets. This study illustrates importance auditing datasets for biases, with particular focus on disaster risk management applications, as lack local data may direct humanitarian actors to utilize global building estimate damage and distribution aid efforts. It important ensure that there are no against representation vulnerable populations they...

Geodata, geographical information science (GISc), and GeoAI (geo-intelligence workflows) play an increasingly important role in predictive disaster risk reduction management (DRRM), aiding decision-makers determining where when to allocate resources. There have been discussions on the ethical pitfalls of these systems context DRRM because documented cases biases AI other socio-technical systems. However, none expound how audit geo-intelligence workflows for from data collection, processing,...

Over the past two decades, humanitarian conduct has been drifting away from classical paradigm. This drift is caused by blurring of boundaries between development aid and humanitarianism increasing reliance on digital technologies data. New humanitarianism, especially in form disaster risk reduction, involved government authorities plans to strengthen their capacity deal with disasters. Digital now enrolls remote data analytics: GIS capacity, local information management experts, volunteers....

Abstract In this study, we present a machine‐learning model capable of predicting food insecurity in the Horn Africa, which is one most vulnerable regions worldwide. The region has frequently been affected by severe droughts and crises over last several decades, will likely increase future. Therefore, exploring novel methods increasing early warning capabilities vital importance to reducing food‐insecurity risk. We XGBoost predict food‐security up 12 months advance. used >20 data sets...

Abstract Damage models for natural hazards are used decision making on reducing and transferring risk. The damage estimates from these depend many variables their complex sometimes nonlinear relationships with the damage. In recent years, data‐driven modeling techniques have been to capture those relationships. available data build such often limited. Therefore, in practice it is usually necessary transfer a different context. this article, we show that implies samples model not fully...

The global shift within disaster governance from response to preparedness and risk reduction includes the emergency of novel Early Warning Systems such as impact based forecasting forecast-based financing. In this new paradigm, funds usually reserved for can be released before a happens when an impact-based forecast—i.e., expected humanitarian result forecasted weather—reaches predefined danger level. development these models are promising, but they also come with significant implementation...

Abstract. Tropical cyclones (TCs) produce strong winds and heavy rains accompanied by consecutive events such as landslides storm surges, resulting in losses of lives livelihoods, particularly regions with high socioeconomic vulnerability. To proactively mitigate the impacts TCs, humanitarian actors implement anticipatory action. In this work, we build upon an existing action for Philippines, which uses impact-based forecasting model housing damage based on eXtreme Gradient Boosting...

Reporting on the Sustainable Development Goals (SDGs) is complex given wide variety of governmental and NGO actors involved in development projects as well increased number targets indicators. However, data indicators must be collected regularly, a robust manner, comparable across but also within countries at different administrative disaggregated levels for adequate decision making to take place. Traditional census household survey not enough. The increase Small Big Data streams have...

It is often taken as given that community-based disaster risk reduction (CBDRR) serves a mechanism for the inclusion of local knowledge (LK) in (DRR). In this paper, through in-depth qualitative analysis empirical data from Malawi, we investigate extent to which CBDRR practice really takes into account LK. This research argues LK underutilised and finds current provides limited opportunity LK, due five prime obstacles: i) approach community participation, ii) financial constraints capacity...

Disaster risk financing has seen a transformative approach through Anticipatory Action (AA), designed to reduce shock and impact of multiple hazards on vulnerable population. The core AA relies pre-agreed triggering mechanisms, that are built around impact-based forecasts (IBF) tailored local contexts, determining when, where, what interventions required. While numerous humanitarian actors have adopted in the recent years, they often work silos, employing varying definitions, methodologies,...

Flood risk can be reduced at various stages of the disaster management cycle. Traditionally, permanent infrastructure is used for flood prevention, while residual managed with emergency measures that are triggered by forecasts. Advances in forecasting hold promise a more prominent role to forecast-based measures. In this study, we present methodology compares flood-prevention On basis methodology, demonstrate how operational decision-makers select between acting against frequent low-impact,...