Global Flood Analysis
Global modeling of seasonal mortality rates from river floods (Earth’s Future)
In this paper published in Earth’s Future, the authors specifically dissect the seasonal response of the global flood events (including the peak timing, peak distribution, flood impact, mortality rate) with the aid of global hydrological model LISFLOOD (GloFAS-Reanalysis dataset v3.0: https://data.jrc.ec.europa.eu/collection/id-00288)
From the analysis, they not only represent major watersheds in the globe for the pure flood generation, but also sheding lights on political boundaries (i.e. countries) for flood assessment since measures have been taken differently with countries.
It is claimed as the first paper to study on global seasonal flood analysis.
In this paper, the authors presented six challenges for global flood risk mapping and came with their solutions:
-
reliable terrain data
-
extreme flow generation either by hydrologic model or regional flood frequency analysis
-
global river networks
They approached the river width by using web-survey, and the river depth is estimated based on 1 or 2 year return period discharge as bankfull discharge. With bankful discharge, channel width, and slope, the channel depth could be approached by manning’s equation.
- Flood defenses
This is more challenging than the other because the structure is way smaller than the modelling grid-cell size
- Computational Hydraulic Engine
“A novel simplified implementation of shallow water equations yielded an glgorithm for which the minimum stable time step scales linearly with decreasing grid size, rather than quadratically as had been the case with previous diffusion wave formulations”
“The study determined that inclusion of both the channel network and floodplain was essential, and that inclusion of the smaller subgrid channels on the floodplain yielded significantly increased simulation accuracy in terms of water level, wave propagation speed, and inundation extent”
- Automation Framework
Global Reconstruction of Naturalized River Flows at 2.94 Million Reaches
A realisation of 35-yr global stream discharge with VIC land surface model and RAPID vector routing scheme. Over 2 million reaches are simulated globally.
For the LSM, they used the VIC model by Liang et al., calibrated based on global streamflow characteristics (generated by Beck et al., 2015).
They used a sparse PDF matching to overcome model calibration or bias correction errors for postprocessing.
Global Maps of Streamflow Characteristics Based on Observations from Several Thousand Catchments
Analysis of global streamflow data, based on flow characteristics.
In this paper, the authors factored a few terms to describe the relationship between flood and precipitation at monthly scales. They investigated floodiness (occurence above certain threshold), duration floodiness, max discharge within a 30 year time span.
- precipitation and discharge appear to be relatively well correlated; however, it is not correlated as good as discharge.
- They disencouraged approaches that simply use precipitation as proxy to interpret floods.
A global streamflow reanalysis for 1980–2018 (J, Hydrol.)
Refined and extended results for Global Flood Awareness System (GloFAS).
In this paper, they described detailed calibration process to produce the 38 year reanalysis results at daily step.
The authors used 11 global hydrologic models to feed into CaMaFlood, to derive global flood extent based on three climate reanalyses dataset. They come up with an idea to threshold low-level flood by eliminating 2-year or below flood due to flood protection measures.
- All models exhibit overestimation of flood extent, yet with flood protection, all models underestimate
- Models in geneneral, are consistent and do not vary significantly across structures and forcing data.
Satellite imaging reveals increased proportion of population exposed to floods (Nature)
This article extracted near 1000 flood events from DFO event reports and cloud-free MODIS imagery with which the authors analyzed flood exposures to global residents using high-resolution settlement dataset. A nice structured paper to alert local communities in the face of climate change.
Predicting flood damage probability across the conterminous United States
As we all know, FEMA has mapped US flood plains at relatively high resolution for different frequencies, but many regions are unmapped due to costly computational resources. This study actually reported that over 68% of the flood events were outside of the mapped flood-plains, and 16.2% were in unmapped flood-plains. This should araise people’s awareness that either past floodplain data is outdated or incomplete.
Use of Hydrological Models in Global Stochastic Flood Modeling
This study compared two global hydrological models - PCR-GLOBWB and CaMa-Flood to investigate global flood modeling. They used the co-occurrence matrix to understand the model predictability compared to in-situ stream gauges.
Regional Flood analysis
Combined modelling of US fluvial, pluvial and coastal flood hazard under current and future climates
In this paper, the authors updated the current US continental flood hazard maps with assigned frequency under different scenarios (historical, present, and projection). They highlighted the changes in hydrography data, surface water profile, extreme event magnitudes, downscaling using hi-res. terrain data, and incorporated local interventions.
They detailed how the boundary conditions are derived for the consideration of fluvial (IDF forced hydrologic simulation) and coastal flooding (tide gauge data), while for pluvial flooding, they simply uses rain-on-grid simulations at each 30 m grid.
They indicate the ensemble prediction is important to better quantify the uncertainties and more model validation efforts are needed.
Implications of Using Global Digital Elevation Models for Flood Risk Analysis in Cities
In this paper, the authors compared six globbaly available DEM data, and assessed how the DEM coupld affect urban flood forecast. They found OS Terrain 50 has the best statistics when validated against high water marks and flood extent derived from photos, videos.
They finally discussed that the outlet water level data in water channles does not nessarily need to be accurate in impact studies if peak floodplain extent is accurate enough.
How probable is widespread flooding in the United States?
In this paper, the authors used a stochastic streamflow generator built upon 38 years USGS streamflow data conditioned on main 18 catchments to investigate the spatial distribution of flood events in dependence of flood susceptibility.
They found a strong dependence of flood susceptibility and seasonality in parts of the US, especially western US. Some catchments governed by intermittent streamflow however exhibits less seasonality. Catchments with snowmelting dominant streamflow generation has high susceptibility in spring and winter.
In this paper, the authors firstly evaluate the model sensitivity with respect to parameters, inflow and topography in terms of flood dynamics. They not only simulated based on bulk flow, but also spatiotemporal variation of flood dynamics. It is worth trying to conduct similar research for CREST-iMAP model parameters or probabilistic QPEs.
A COMPREHENSIVE DATABASE OF FLOOD EVENTS IN THE CONTIGUOUS UNITED STATES FROM 2002 TO 2013
The authors constructed US flood database based on gauge data and precipitation data to separate baseflow from flood events.
FLood seasonality
On the seasonality of flooding across the continental United States
In this paper, author comprehensively discussed US flood seasonality from meteorological point of view as well as anthropogenic effects such as urbanization and flow regulation. But he claims climatology plays a main role, and urbanisation does not contribute much to the flood seasonality shifts.
Peak Runoff Timing Is Linked to Global Warming Trajectories (Earths Future
The authors analyzed CMIP5 models using Bayesian Mean Ensenble approach to extract runoff time series over the United States and compared the peak timings. They found In snowmelt-dominated areas, annual maxima are projected to shift to earlier dates due to the corresponding changes in snow accumulation timing. For regions in which the occurrence of springtime extreme soil wetness shifts to later time, we find that peak annual runoff is also projected to be delayed.
Northeast US
The Seasonal Nature of Extreme Hydrological Events in the Northeastern United States
A long-term trend analysis is done for precipitation and streamflow separatly from the perspective of large scale atmospheric circulation. The authors ablate warm season changes from lumped one, and they found most robust and significant changes are present for warm-season precipitation and streamflow, although precipitation extremes are more obvious than streamflow.
Central US
The changing nature of flooding across the central United States (Nat. Climate Change)
In this paper, the authors investigated the flood changes (i.e., flood magnitude and flood frequency) with respect to rainfall changes in the Central US. They attribute most of the flood changes to extreme rainfall changes.
Eastern US
Spatial heterogeneities in flood peak distributions due to orographic precipitation mechanisms in mountainous terrain, coastal circulations near land–ocean boundaries, and urbanization impacts on regional climate are central features of flood peak distributions for the eastern United States, and for many other settings around the world. Orographic mechanisms in the eastern United States associated with heavy rainfall from tropical cyclones, winter–spring extratropical systems, and warm-season thunderstorms are diverse and in many settings poorly understood.
Winter–spring extratropical systems account for a large fraction of eastern U.S. flood peaks, especially in the northeastern and southeastern United States (Fig. 1b). We use the frequency of March–April flood peaks as a surrogate for winter–spring extratropical flood peaks. A combination of snowmelt and rain on snow determines the local maximum in flood peak occurrence in the northeastern United States (Fig. 1b), where March–April peaks account for up to 60% of annual peaks. Organized thunderstorm systems embedded in winter–spring extratropical systems, often associated with severe weather, are important flood agents in the southeastern United States (Fig. 1b). March–April peaks account for more than 50% of annual flood peaks in south Georgia, north Florida, and southeastern Alabama (Fig. 1b).
Warm-season thunderstorm systems are also important flood agents in urban watersheds along the urban megalopolis of the eastern United States
EU
Changing climate shifts timing of European floods (Science)
The authors analyzed nearly 5000 hydrometric stations over EU for the past five decades and found clear patterns of change in flood timing. Warmer temperatures have led to earlier spring snowmelt floods throughout northeastern Europe; delayed winter storms associated with polar warming have led to later winter floods around the North Sea and some sectors of the Mediterranean coast; and earlier soil moisture maxima have led to earlier winter floods in western Europe.
Australia
Changes in Antecedent Soil Moisture Modulate Flood Seasonality in a Changing Climate (WRR)
This article investigated the role of interdependence of annual maximum rainfall, soil moisture, and flood in Australia.
Fig. Map of seasonality and seasonality strength for three variables. Top down: rainfall, soil moisture, and flow.
Flood generation mechanisms
Event-based classification for global study of river flood generating processes (Hydrol. Process)
They classified global streamflow data into six classes: snow/rain flood, snowmelt flood, excess rain, short rain flood, long rain flood, and others.
Field Studies of Hillslope Flow Processes (Dunne 1978)
Dominant flood generating mechanisms across the United States
In this paper, authors proposed four hypothesis to investigate the dominant flood-generating process in the US. They isolated overall processes into extreme rainfall-caused, rainfall-caused, excess-rainfall caused, snow-melt caused. Eventually they clustered US regions with respective flood processes.
The Relative Importance of Different Flood-Generating Mechanisms Across Europe
Three major flood driving mechanisms are identified:
- Extreme precipitation: In extreme precipitation floods, the maximum annual flow results from the largest precipitation event of that year.
- Soil Moisture excess: In soil moisture excess floods, the maximum annual flow is caused by the largest daily soil moisture excess event. It is defined as the daily precipitation amount minus the available soil moisture storage capacity.
- Snowmelt: in snowmelt floods, the maximum annual flow is caused by the largest daily snowmelt or rain-on-snow event of the year.
They also quantified flood seasonality with circular method.
A striking feature is that in Euro, flood dependence on extreme precipitation is weak while it is more or less based off snowmelt and soil moisture excess (most important).
The disconnect between precipitation extremes and flooding helps to contextualize flood trends. For example, observations indicate that annual maximum daily precipitation is increasing (Westra et al., 2013), but such a persistent increase is not found in observed annual maximum floods in Europe (Hall et al., 2014). Similarly, model simulations of European river floods at 1.5, 2, and 3 °C global warming indicate a very weak correlation between changes in precipitation extremes and changes in flood magnitudes (Thober et al., 2018). These observations are not very surprising when we consider that most annual peak floods are in fact not caused by annual peak precipitation. Our results point to the processes (snowmelt and soil moisture excess) that likely better explain flood trends in Europe (Blöschl et al., 2017).
Controls on Flood Trends Across the United States WRR
This study integrates flood generating mechanisms into a framework to analyze flood trends based on stream gauge data. To identify antecedent conditions, they estimated the average concentration time for individual catchments based on their areas.
\[t_c=0.1*A^{0.3}\]They use network analysis to cluster stream gauges based on a similarity metric, consisting of synchronization, generating process similarity, and magnitudes similarity
This article briefly brought up the concern that transforming natural vegetation to rainfed croplands make South America’s sedimentary plains more flood prone, because of increase in local groundwater table. And this increase in waterlogging is not explained by precipitation variability.
Spatial dependence of flood
- Flood identification
- POT: flow rate exceeding 25th percentile of 30-year annual maxima and minimum 10 days time lag between events.
- compile events with all catchments with binary matrix.
- identify regional flood that only affect single catchment.
- Spatial pattern of flood occurence
flood connectedness: co-experience of flood events. F-madogram measures spatial dependence as a function of distance between a pair of stations.
- Pair with hydrometeorologic drivers
Growing Spatial Scales of Synchronous River Flooding in Europe (GRL)
They first introduce the flood synchrony scale, defined as the maximum radius around an individual river gauge within which at least half of the other river gauges also record flooding almost simultaneously.
Joint Trends in Flood Magnitudes and Spatial Extents Across Europe (GRL)
Flood Synchrony scale measure: the maximum distance from a station within which at least 50% of the stations have the annual maximum flood discharge at the same time as the reference station. A 7-day delay is allowed to account for travel time in EU weather and routing system.
They classified flood by its generating mechanisms: 1) convective rainfall; 2) snowfall; 3) rain-on-snow; 4) soil excess rainfall.
An increase/decrease in flood synchrony is shown:
