A climate-conditioned catastrophe risk model for UK flooding

Natural hazards and earth system sciences, Journal Year: 2023, Volume and Issue: 23(2), P. 891 - 908

Published: March 7, 2023

Abstract. We present a transparent and validated climate-conditioned catastrophe flood model for the UK, that simulates pluvial, fluvial coastal risks at 1 arcsec spatial resolution (∼ 20–25 m). Hazard layers 10 different return periods are produced over whole UK historic, 2020, 2030, 2050 2070 conditions using Climate Projections 2018 (UKCP18) climate simulations. From these, monetary losses computed five specific global warming levels above pre-industrial values (0.6, 1.1, 1.8, 2.5 3.3 ∘C). The analysis contains greater level of detail nuance compared to previous work, represents our current best understanding UK's changing risk landscape. Validation against historical national period maps yielded critical success index 0.65 0.76 England Wales, respectively, maximum water Carlisle 2005 were replicated root mean square error (RMSE) 0.41 m without calibration. This skill is similar local modelling with site-specific data. Expected annual damage in 2020 was GBP 730 million, which compares favourably observed value 714 million reported by Association British Insurers. Previous loss estimates based on government data ∼ 3× higher, lie well outside modelled distribution, plausibly centred observations. estimate % probability 6 average 1.1 ∘C warming) those 1990 0.6 warming), this increase can be kept around 8 if all countries' COP26 2030 carbon emission reduction pledges “net zero” commitments implemented full. Implementing only increases 23 values, potentially 37 “worst case” scenario where targets missed sensitivity high.

Language: Английский

---

A web-based analytical urban flood damage and loss estimation framework

Environmental Modelling & Software, Journal Year: 2023, Volume and Issue: 163, P. 105670 - 105670

Published: March 7, 2023

Language: Английский

---

Human alterations of the global floodplains 1992–2019

Scientific Data, Journal Year: 2023, Volume and Issue: 10(1)

Published: July 28, 2023

Floodplains provide critical ecosystem services; however, loss of natural floodplain functions caused by human alterations increase flood risks and lead to massive life property. Despite recent calls for improved protection management, a comprehensive, global-scale assessment quantifying does not exist. We developed the first publicly available global dataset that quantifies in 15 million km

Language: Английский

---

Chapter 2 : Climate Trends. Fifth National Climate Assessment

Published: Jan. 1, 2023

Language: Английский

---

A Framework to Assess Remote Sensing Algorithms for Satellite-Based Flood Index Insurance

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2023, Volume and Issue: 16, P. 2589 - 2604

Published: Jan. 1, 2023

Remotely sensed data has the potential to monitor natural hazards and their consequences on socio-economic systems. However, in much of world, inadequate validation disaster damage make reliable use satellite difficult. We attempt strengthen for one application - flood index insurance which manage largely uninsured losses from floods. Flood is a particularly challenging remote sensing due floods' speed, unpredictability, significant required. propose set criteria assessing algorithm performance provide framework data-poor environments. Within these criteria, we assess several metrics – spatial accuracy compared high-resolution PlanetScope imagery (F1), temporal consistency as river water levels (Spearman's ρ), correlation government (R ² ) that measure performance. With develop Sentinel-1 inundation time series Bangladesh at high (10m) (∼weekly) resolution compare it previous MODIS used insurance. Results show adapted (F1 _avg =0.925, ρ =0.752, R =0.43) significantly outperforms algorithms criteria. Beyond Bangladesh, our proposed can be validate better products other applications places with ground truth data.

Language: Английский

---

Evolution of Floods: From Ancient Times to the Present Times (ca 7600 BC to the Present) and the Future

Land, Journal Year: 2023, Volume and Issue: 12(6), P. 1211 - 1211

Published: June 11, 2023

Floods are one of the most dangerous natural disasters, causing great destruction, damage, and even fatalities worldwide. Flooding is phenomenon a sudden increase or slow in volume water river stream bed as result several possible factors: heavy very long precipitation, melting snowpack, strong winds over water, unusually high tides, tsunamis, failure dams, gages, detention basins, other structures that hold back water. To gain better understanding flooding, it necessary to examine evidence, search for ancient wisdom, compare flood-management practices different regions chronological perspective. This study reviews flood events caused by rising sea levels erratic weather from times present. In addition, this review contemplates concerns about future challenges countermeasures. Thus, presents catalogue past examples order present point departure floods learn lessons preparation incidents including rainfalls, particularly urbanized areas. The results show societies developed multifaceted technologies cope with many them still usable now may represent solutions measures counter changing increasingly more

Language: Английский

---

Application of a hybrid algorithm of LSTM and Transformer based on random search optimization for improving rainfall-runoff simulation

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: May 16, 2024

Flood forecasting using traditional physical hydrology models requires consideration of multiple complex processes including the spatio-temporal distribution rainfall, spatial heterogeneity watershed sub-surface characteristics, and runoff generation routing behaviours. Data-driven offer novel solutions to these challenges, though they are hindered by difficulties in hyperparameter selection a decline prediction stability as lead time extends. This study introduces hybrid model, RS-LSTM-Transformer, which combines Random Search (RS), Long Short-Term Memory networks (LSTM), Transformer architecture. Applied typical Jingle middle reaches Yellow River, this model utilises rainfall data from basin sites simulate flood processes, its outcomes compared against those RS-LSTM, RS-Transformer, RS-BP, RS-MLP models. It was evaluated Nash-Sutcliffe Efficiency Coefficient (NSE), Root Mean Square Error (RMSE), Absolute (MAE), Bias percentage metrics. At 1-h during calibration validation, RS-LSTM-Transformer achieved NSE, RMSE, MAE, values 0.970, 14.001m

Language: Английский

---

Urban flood susceptibility mapping using frequency ratio and multiple decision tree-based machine learning models

Natural Hazards, Journal Year: 2024, Volume and Issue: 120(11), P. 10365 - 10393

Published: April 23, 2024

Language: Английский

---

National‐Scale Flood Hazard Data Unfit for Urban Risk Management

Earth s Future, Journal Year: 2024, Volume and Issue: 12(7)

Published: July 1, 2024

Abstract Extreme flooding events are becoming more frequent and costly, impacts have been concentrated in cities where exposure vulnerability both heightened. To manage risks, governments, the private sector, households now rely on flood hazard data from national‐scale models that lack accuracy urban areas due to unresolved drainage processes infrastructure. Here we assess uncertainties of First Street Foundation (FSF) data, available across U.S., using a new model (PRIMo‐Drain) resolves infrastructure fine resolution dynamics. Using case Los Angeles, California, find FSF PRIMo‐Drain estimates population property value exposed 1%‐ 5%‐annual‐chance hazards diverge at finer scales governance, for example, by 4‐ 18‐fold municipal scale. often predict opposite patterns inequality social groups (e.g., Black, White, Disadvantaged). Further, county scale, compute Model Agreement Index only 24%—a ∼1 4 chance agreeing upon which properties risk. Collectively, these differences point limited capacity confidently municipalities, groups, individual risk within areas. These results caution present may misinform strategies lead maladaptation, underscoring importance refined validated models.

Language: Английский

---

Large floods drive changes in cause-specific mortality in the United States

Nature Medicine, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 3, 2025

Flooding greatly endangers public health and is an urgent concern as rapid population growth in flood-prone regions more extreme weather events will increase the number of people at risk. However, exhaustive analysis mortality following floods has not been conducted. Here we used 35.6 million complete death records over 18 years (2001–2018) from National Center for Health Statistics United States, highly resolved flood exposure data a Bayesian conditional quasi-Poisson model to estimate association flooding with monthly county-level rates cancers, cardiovascular diseases, infectious parasitic injuries, neuropsychiatric conditions respiratory diseases up 3 months after flood. During month flooding, very severe heavy rain-related were associated increased disease (3.2%; 95% credible interval (CrI): 0.1%, 6.2%) (2.1%; CrI: 1.3%, 3.0%) tropical cyclone-related injury (15.3%; 12.4%, 18.1%). increases rate higher those ≥65 old (24.9; 20.0%, 29.8%) than aged <65 (10.2%; 6.6%, 13.8%) females (21.2%; 16.3%, 26.1%) males (12.6%; 9.1%,16.1%). Effective responses are critical now projected severity driven by climate change. Analyses two decades States reported greater injuries exposures events, particularly caused cyclones rain.

Language: Английский

---