Flood hazard potential reveals global floodplain settlement patterns

Nature Communications, Journal Year: 2023, Volume and Issue: 14(1)

Published: May 16, 2023

Flooding is one of the most common natural hazards, causing disastrous impacts worldwide. Stress-testing global human-Earth system to understand sensitivity floodplains and population exposure a range plausible conditions strategy identify where future changes flooding or might be critical. This study presents analysis inundated areas varying flood event magnitudes globally for 1.2 million river reaches. Here we show that topography drainage correlate with sensitivities as well societal behaviour. We find clear settlement patterns in which sensitive frequent, low magnitude events, reveal evenly distributed across hazard zones, suggesting people have adapted this risk. In contrast, extreme events tendency populations densely settled these rarely flooded being significant danger from potentially increasing given climate change.

Language: Английский

---

Characterizing uncertainty in process-based hydraulic modeling, exemplified in a semiarid Inner Mongolia steppe

Geoderma, Journal Year: 2023, Volume and Issue: 440, P. 116713 - 116713

Published: Nov. 16, 2023

Assessing root sources of three uncertainties – parameterization soil hydraulic characteristics, boundary conditions, and estimation source/sink terms is a significant challenge in water transport modeling. This study aims to evaluate the uncertainty each widely-used parameter methods affecting plot-scale dynamics. The employs HYDRUS, process-based hydrologic model, incorporate these compare model predictions measured values semiarid Inner Mongolia steppe, China. Soil parameters are determined using two direct (laboratory-derived approach evaporation method) one indirect method (neural network). While generally simulates moisture dynamics, performed better, especially under dry conditions. suggests that measuring intensity properties, such as unsaturated conductivity, with crucial for reasonable simulation. also demonstrates impact different applied conditions on simulated moisture, specifically partitioning reference FAO evapotranspiration via (soil fraction cover) (leaf area index crop height). cover reflected better Additionally, compares uptake function growth constant depth referenced grass pasture, finds no difference among them. Comparing predicting concludes input more sensitive than or representation function. Our highlights properties can reflect effects land use change, compaction, field transports.

Language: Английский

---

Temporal Fusion Transformers for streamflow Prediction: Value of combining attention with recurrence

Journal of Hydrology, Journal Year: 2024, Volume and Issue: 637, P. 131301 - 131301

Published: May 9, 2024

Language: Английский

---

The potential to reduce uncertainty in regional runoff projections from climate models

Nature Climate Change, Journal Year: 2019, Volume and Issue: 9(12), P. 926 - 933

Published: Nov. 26, 2019

Language: Английский

---

Effects of univariate and multivariate bias correction on hydrological impact projections in alpine catchments

Hydrology and earth system sciences, Journal Year: 2019, Volume and Issue: 23(3), P. 1339 - 1354

Published: March 11, 2019

Abstract. Alpine catchments show a high sensitivity to climate variation as they include the elevation range of snow line. Therefore, correct representation variables and their interdependence is crucial when describing or predicting hydrological processes. When using model simulations in impact studies, forcing meteorological data are usually downscaled bias corrected, most often by univariate approaches such quantile mapping individual variables, neglecting relationships that exist between variables. In this study we test hypothesis explicit consideration relation air temperature precipitation will affect modelling snow-dominated mountain environment. Glacio-hydrological were performed for two partly glacierized alpine recently developed multivariate correction method post-process EURO-CORDEX regional outputs 1976 2099. These compared those obtained common correction. As both methods each variable's distribution same way, marginal distributions no differences. Yet, regarding temperature, clear differences notable studied catchments. Simultaneous based on approach led more below temperatures 0 ∘C therefore simulated snowfall than with approach. This difference translated considerable consequences responses The bias-correction-forced showed distinctly different results projected cover characteristics, snowmelt-driven streamflow components, expected glacier disappearance dates. all aspects – fraction above ∘C, water equivalents, volumes, regime resulting from multivariate-corrected corresponded better reference Differences total due may be considered negligible given generally large spread projections, but systematic seasonally delayed components snowmelt particular matter planning perspective. While does not allow conclusive evidence preferable, it clearly demonstrates incorporating ignoring inter-variable can conclusions drawn change studies environments.

Language: Английский

---

Transferability of hydrological models and ensemble averaging methods between contrasting climatic periods

Water Resources Research, Journal Year: 2016, Volume and Issue: 52(10), P. 8343 - 8373

Published: Oct. 1, 2016

Understanding hydrological model predictive capabilities under contrasting climate conditions enables more robust decision making. Using Differential Split Sample Testing (DSST), we analyze the performance of six models for 37 Irish catchments unlike those used training. Additionally, consider four ensemble averaging techniques when examining interperiod transferability. DSST is conducted using 2/3 year noncontinuous blocks (i) wettest/driest years on record based precipitation totals and (ii) with a more/less pronounced seasonal regime. Model transferability between regimes was found to vary depending testing scenario, catchment, evaluation criteria considered. As expected, average outperformed most individual members. However, differed considerably in number times they surpassed best member. Bayesian Averaging (BMA) Granger-Ramanathan (GRA) method were outperform simple arithmetic mean (SAM) Akaike Information Criteria (AICA). Here GRA performed better than 51%–86% cases (according Nash-Sutcliffe criterion). When assessing skill change recommend setting up select available analogues expected annual conditions; applying multiple criteria; (iii) diverse set catchments; (iv) multimodel conjunction an appropriate technique. Given computational efficiency relative BMA, former recommended as preferred technique assessment.

Language: Английский

---

Future shifts in extreme flow regimes in Alpine regions

Hydrology and earth system sciences, Journal Year: 2019, Volume and Issue: 23(11), P. 4471 - 4489

Published: Oct. 30, 2019

Extreme low and high flows can have negative economic, social, ecological effects are expected to become more severe in many regions due climate change. Besides flows, the whole flow regime, i.e., annual hydrograph comprised of monthly mean is subject changes. Knowledge on future changes regimes important since contain information both extremes conditions prior dry wet seasons. Changes individual low- high-flow characteristics as well under been thoroughly studied. In contrast, little known about extreme regimes. We here propose two methods for estimation apply them simulated discharge time series Switzerland. The first method relies frequency analysis performed duration curves. second approach performs sums a large set stochastically generated hydrographs. Both approaches were found produce similar 100-year regime estimates when applied data 19 hydrological Our results show that rainfall-dominated distinct from those melt-dominated regions. regions, minimum low-flow decreases by up 50 %, whilst reduction 25 % maximum increases %. point other direction than discharges increase 100 decrease less respectively. findings provide guidance water resource planning management valuable basis impact studies. Highlights Estimation using curves will change conditions. but

Language: Английский

---

Variability of orographic enhancement of precipitation in the Alpine region

Scientific Reports, Journal Year: 2019, Volume and Issue: 9(1)

Published: Sept. 16, 2019

Abstract Climate change impacts are non uniformly distributed over the globe. Mountains have a peculiar response to large scale variations, documented by elevation gradients of surface temperature increase observed many mountain ranges in last decades. Significant changes precipitation expected changing climate and orographic effects important determining amount rainfall at given location. It thus becomes particularly understand how responds global warming anthropogenic forcing. Here, using rain gauge dataset European Alpine region, we show that distribution annual among lowlands mountains has varied time, with an high elevations compared low starting mid 20 century peaking 1980s. The simultaneous peak aerosol load is discussed as possible source for this interdecadal change. These results provide new insights further our understanding improve predictions anthropic on precipitations, which fundamental water security management.

Language: Английский

---

Changing climate drives future streamflow declines and challenges in meeting water demand across the southwestern United States

Journal of Hydrology X, Journal Year: 2021, Volume and Issue: 11, P. 100074 - 100074

Published: Jan. 27, 2021

Society and the environment in arid southwestern United States depend on reliable water availability, yet current use outpaces supply. Water demand is projected to grow future climate change expected reduce To adapt, managers need robust estimates of regional supply support management decisions. address this need, we estimate streamflow seven resource regions U.S. using a new SPAtially Referenced Regressions On Watershed attributes (SPARROW) model. We present projections corresponding input data from models two greenhouse gas Representative Concentration Pathways (RCP4.5 8.5) for three, thirty-year intervals centered 2030s, 2050s, 2080s, historical thirty year interval 1990s. Across regions, about half RCP4.5 (51%) thirds RCP8.5 (67%) indicate decreases 2080s relative period. Models project maximum 36–80% all periods RCPs streamflow, up 20–45% at sites along Colorado River used measuring compliance with interstate international agreements. Headwaters are experience greatest declines, substantial downstream implications. Among these estimates, streamflows forced tend be lower than those RCP4.5. Not models, times, widespread declines. The most ubiquitous increases occur 2030s under Later time enhanced forcings smaller increase accumulated streamflows, suggesting that limiting or reducing concentrations could availability. Although some possible promising, modest spatially limited later still unlikely sufficient meet demand. These results inform likelihood agreement compliance, developing strategies balance

Language: Английский

---

A note on leveraging synergy in multiple meteorological data sets with deep learning for rainfall–runoff modeling

Hydrology and earth system sciences, Journal Year: 2021, Volume and Issue: 25(5), P. 2685 - 2703

Published: May 20, 2021

Abstract. A deep learning rainfall–runoff model can take multiple meteorological forcing products as input and learn to combine them in spatially temporally dynamic ways. This is demonstrated with Long Short-Term Memory networks (LSTMs) trained over basins the continental US, using Catchment Attributes Meteorological data set for Large Sample Studies (CAMELS). Using from different (North American Land Data Assimilation System, NLDAS, Maurer, Daymet) a single LSTM significantly improved simulation accuracy relative only individual products. sensitivity analysis showed that combines precipitation ways, depending on location, also ways of parts hydrograph.

Language: Английский

---