Search

The recent surge in the frequency, severity, and extent of wildfires, along with the increased risk of wildfire-induced flooding, highlights the need to quantify the potential impacts of wildfires on peak flood flows. However, supporting wildfire impact assessments with imprecise models can be challenging due to the detailed information typically required about the severity and extent of wildfires, degree of dynamic forest recovery, and a lack of postburn flow data. Moreover, making reasonable assumptions about wildfire impacts becomes difficult. To address this challenge, we propose a novel methodology for screening wildfire impact scenarios on peak flood flows in regions with limited data before a wildfire has occurred. This methodology includes prefire process-based hydrological modeling, sequentially screening short wildfire impacts, and flood frequency analysis. As a proof of concept, the current strategy has been applied to four fire-prone watersheds in Canada. Unburned and worst-burn scenarios were generated and compared to quantify changes in peak flood flows and flood frequency curves. The results indicated that annual peak flows and flood frequency curves experienced an increase in the short-term worst-burn scenario across all four watersheds. The proposed screening methodology estimates the upper limits of postfire peak flood flows, offering insights into which watersheds may be disproportionately impacted by a wildfire regime. This model outputs can be seamlessly integrated into a risk management framework to inform wildfire management decisions aimed at hazard prevention and risk reduction. Practical Applications: This study introduces a groundbreaking methodology for screening the potential impact of wildfires on peak flood flows, even in regions with limited data and before a fire occurs. By using prefire hydrological models, simulating short-term wildfire effects, and analyzing flood frequency, this approach allows for early identification of watersheds that are highly vulnerable to postfire flooding. Specifically, it distinguishes between watersheds that are strongly or weakly affected by wildfire in the worst-case scenario, where little information is available about the extent or severity of the burn. This methodology may eventually be enhanced with additional data on burn severity for specific forest types, yet it currently provides a critical tool for categorizing watershed vulnerability to wildfire-related flooding. [ABSTRACT FROM AUTHOR]

Traditional flood frequency analysis assumes stationary conditions (i.e. the mean and other statistical properties are unchanging) prevail in the physical and climatological element driving the phenomenon. With climate change and rapid landcover change, this assumption must be reviewed, and new approaches considering nonstationarity may need to be adopted. Long-term rainfall and land cover data were used to reconstruct historical streamflow in three urban watersheds using deterministic and stochastic techniques. The streamflow models were developed with static and time-evolving built-up land cover area to mimic the effect of land cover change due to urbanization. Annual flood maximum series were developed from each streamflow data set and were tested for trends. The models with time-evolving built-up landcover area (deterministic models) and those with urbanization as co-predictors (stochastic models) were able to generate continuous streamflow time series that yielded flood extremes exhibiting nonstationarity. The annual flood maxima were fitted onto stationary and nonstationary Generalized Extreme Value distribution models using Bayesian approach and successively tested for goodness-of-fit and parsimony. All the annual flood series from both deterministic and stochastic models satisfactorily fit both the stationary and nonstationary Generalized Extreme Value distribution, with the stationary models exhibiting better fit for streamflow models of watersheds with static urbanization scenarios; and the nonstationary models exhibiting better fit for streamflow models of watersheds with evolving urbanization scenarios. In terms of parsimony, the stochastically generated flood models are better than those developed from deterministic models as evidenced by the lower Akaike Information Criterion and Bayesian Information Criterion values for all watersheds. The probability of exceedance of floods through some threshold magnitude increases under nonstationary conditions. [ABSTRACT FROM AUTHOR]

The Kopili River, which is the largest south-bank tributary of the river Brahmaputra, is primarily responsible for the frequent floods in the Morigaon and Nagaon districts of Assam, a northeastern state of India. The area of Dharamtul in Morigaon district is particularly susceptible to floods every year. Therefore, the current study aims to examine the hydrological characteristics of the Kopili River at the Dharamtul site so that insights can be drawn regarding the annual flooding patterns in the region. In this research, historical time series data has been utilized to create a comprehensive hydrological database. The database includes information such as peak discharge, design discharge for different return periods (10 years, 25 years, 50 years, and 100 years), stage-discharge rating curve, return period, and annual hydrograph for the Kopili river at the selected site, which is also a designated gauge site for the Water Resource Department of the Government of Assam. The findings of this research will be valuable in formulating flood models and implementing flood mitigation measures in the region in the long run. Additionally, it will pave the way for further hydrological studies in the area. [ABSTRACT FROM AUTHOR]

The Lower Kosi Basin (LKB) in North Bihar is highly prone to floods and is influenced by upstream hydrology. A flood susceptibility index has been modelled by integrating eleven flood conditioning parameters (precipitation, elevation, slope, drainage density, distance from the river, ruggedness index, topographic wetness index, stream power index, curvature, normalized difference vegetation index, land use and land cover) derived from the satellite data, using a weighted linear summation model. The study uses Sentinel-1 synthetic aperture radar data to estimate flood frequency over a temporal scale of 2016–2020. The flood frequency was used to validate the flood susceptibility derived using multi-criteria decision making methods combined with geographical information system (MCDM-GIS). The study shows that ~ 66% of the area in LKB is susceptible to high to moderate flooding while the remaining ~ 34% is falls in the low flooding category. 15.24% of the area has high frequency (> 3 flood occurrences) of the flood, 9.66% has moderate (2 flood occurrences) and 9.72% of the area faced one-time flood during five years of period (2016–2020). The accuracy of MCDM-GIS derived flood susceptibility map was assessed using area under curve, confusion matrix, precision, recall, F1 score, weighted F1 score and overall accuracy. [ABSTRACT FROM AUTHOR]

Study region: Taihu Basin, China Study focus: Floods threaten humans, the environment, economic activity, and infrastructure. In this study, a new trend test and flood-frequency methods were adopted to detect extreme floods and their distributions based on flood-event identification. To fully understand the phased process of the influence of human activities on extreme hydrological processes, 12 copula functions were employed creatively in combined static and dynamic time-varying correlation aspects between extreme precipitation and floods. New hydrological insights for the region: Although both significant and insignificant increasing trends of the annual maximum water level in all three hydrological districts were examined, the periodic oscillations of all the stations were similar. Thus, it was significant to fully detect the periodical variation of floods. Extreme floods occurred mainly in the 1990s, as measured by frequency estimates. Generally, the nonstationary response relationship between heavy rain and an extreme water level was gradually strengthened; that is, a certain magnitude precipitation seemed to induce a greater-intensity flood event as time passed. Through the identification of historical flood events and the analysis of the rise and fall processes of floods, we found that the main reason for variation in the response relationship was the increase in the water level before the rising stage, rather than the water level rising in the Taihu Basin. Our study findings further existing knowledge on the regional flood-control design standard and can ensure the coexistence of humans and water systems in the future.

Flood frequency in river source regions is significantly affected by rainfall and snowmelt as part of climatic changes. A traditional univariate flood frequency analysis cannot reflect the complexity of floods, and when used in isolation, it can only underestimate flood risk. For effective flood prevention and mitigation, it is essential to consider the combined effects of precipitation and snowmelt. Copula functions can effectively quantify the joint distribution relationship between floods and their associated variables without restrictions on their distribution characteristics. This study uses copula functions to consider a multivariate probability distribution model of flood peak flow (Q) with cumulative snowmelt (CSm) and cumulative precipitation (CPr) for the Hutubi River basin located in northern Xinjiang, China. The joint frequencies of rainfall and snowmelt floods are predicted using copula models based on the Coupled Model Intercomparison Project Phase 6 data. The results show that Q has a significant positive correlation with 24-d CSm (r = 0.559, p = 0.002) and 23-d CPr (r = 0.965, p

Flood estimates based on stationary flood frequency models are commonly used as inputs to flood hazard mapping. However, changing flood characteristics caused by climate change necessitate more accurate assessments of the probabilities of rare flood events. This study aims to develop a flood hazard map based on the nonstationary flood frequency using a generalized extreme value distribution model for the Becho floodplain in the upper Awash River basin. The distributional location parameter was modeled as a function of rainfall amount of different durations, annual total precipitation from wet days, yearly mean maximum temperature and time as covariates. The one-dimensional Hydrological Engineering Center River Analysis System (HEC-RAS) hydraulic model with steady flow analysis was used to generate flood hazard map input, depth and velocity, and inundation extent for different return periods. The result indicated that the model as a function of rainfall, such as monthly rainfall (August) and annual wet day precipitation, provided the best fit to the observed hydrological data. Rainfall as a covariate can explain the variation in the peak flood series. The developed hazard map based on depth alone and the combination of depth and velocity thresholds resulted in more than 70% of the floodplain area being classified as a high hazard zone under 2, 25, 50, and 100-years return periods. The current study assists water resource managers in considering changing environmental factors and an alternative flood frequency model for developing flood hazard management and mitigation strategies. [ABSTRACT FROM AUTHOR]

Non-stationary methods of flood frequency analysis are widespread in research but rarely implemented by practitioners. One reason may be that research papers on non-stationary statistical models tend to focus on model fitting rather than extracting the sort of results needed by designers and decision makers. It can be difficult to extract useful results from non-stationary models that include stochastic covariates for which the value in any future year is unknown. We explore the motivation for including such covariates, whether on their own or in addition to a covariate based on time. We set out a method for expressing the results of non-stationary models as an integrated flow estimate, which removes the dependence on the covariates. This can be defined either for a particular year or over a longer period of time. The methods are illustrated by application to a set of 375 river gauges across England and Wales. We find annual rainfall to be a useful covariate at many gauges, sometimes in conjunction with a time-based covariate. For estimating flood frequency in future conditions, we advocate exploring hybrid approaches that combine the best attributes of non-stationary statistical models and simulation models that can represent changes in climate and river catchments. HIGHLIGHTS We explore why and how to include physical variables as covariates in statistical models of flood frequency.; We develop and illustrate methods for extracting flow estimates from such models so that practitioners can obtain useful results.; Practitioners now have tools and guidance to apply non-stationary methods for flood management in England.;

Copyright of China Rural Water & Hydropower is the property of China Rural Water & Hydropower Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Hydrology and meteorological data from relatively undisturbed watersheds aid in identifying effects on ecosystem services, tracking hydroclimatic trends, and reducing model uncertainties. Sustainable forest, water, and infrastructure management depends on assessing the impacts of extreme events and land use change on flooding, droughts, and biogeochemical processes. For example, global climate models predict more frequent high-intensity storms and longer dry periods for the southeastern USA. We summarized 17 years (2005–2021) of hydrometeorological data recorded in the 52 km2, third-order Turkey Creek watershed at the Santee Experimental Forest (SEF), Southeastern Coastal Plain, USA. This is a non-tidal headwater system of the Charleston Harbor estuary. The study period included a wide range of weather conditions; annual precipitation (P) and potential evapotranspiration (PET) ranged from 994 mm and 1212 mm in 2007 to 2243 mm and 1063 in 2015, respectively. The annual runoff coefficient (ROC) varied from 0.09 in 2007 (with water table (WT) as deep as 2.4 m below surface) to 0.52 in 2015 (with frequently ponded WT conditions), with an average of 0.22. Although the average P (1470 mm) was 11% higher than the historic 1964–1976 average (1320 mm), no significant (α= 0.05) trend was found in the annual P (p = 0.11), ROC (p = 0.17) or runoff (p = 0.27). Runoff occurred on 76.4% of all days in the study period, exceeding 20 mm/day for 1.25% of all days, mostly due to intense storms in the summer and lower ET demand in the winter. No-flow conditions were common during most of the summer growing season. WT recharge occurred during water-surplus conditions, and storm-event base flow contributed 23–47% of the total runoff as estimated using a hydrograph separation method. Storm-event peak discharge in the Turkey Creek was dominated by shallow subsurface runoff and was correlated with 48 h precipitation totals. Estimated precipitation intensity–duration–frequency and flood frequency relationships were found to be larger than those found by NOAA for the 1893–2002 period (for durations ≥ 3 h), and by USGS regional frequencies (for ≥10-year return intervals), respectively, for the same location. We recommend an integrated analysis of these data together with available water quality data to (1) assess the impacts of rising tides on the hydroperiod and biogeochemical processes in riparian forests of the estuary headwaters, (2) validate rainfall–runoff models including watershed scale models to assess land use and climate change on hydrology and water quality, and (3) inform watershed restoration goals, strategies, and infrastructure design in coastal watersheds. [ABSTRACT FROM AUTHOR]

This study evaluates five bias correction and statistical downscaling (BCSD) techniques for daily precipitation and examines their impacts on the projected changes in flood extremes (i.e., 1%, 0.5%, and 0.2% floods). We use climate model outputs from the Coupled Model Intercomparison Project Phase 6 (CMIP6) to conduct hydrologic simulations across watersheds in Iowa and determine historical and future flood extreme estimates based on generalized extreme value distribution fitting. Projected changes in these extremes are examined with respect to four Shared Socioeconomic Pathways (SSPs) alongside five BCSD techniques. We find the magnitude of the estimates of future annual exceedance probabilities (AEPs) are expected to increase under all SSPs, especially for the emission scenarios with higher greenhouse gases concentrations (i.e., SSP370 and SSP585). Our results also suggest the choice of BCSD impacts the magnitude of the projected changes, with the SSPs that play a more limited role compared to the choice of downscaling method. The variability in projected flood changes across Iowa is similar across the downscaling technique but increases as the AEP increases. Our findings provide insights into the impact of downscaling techniques on flood extremes' projections and useful information for climate planning across the state. Plain Language Summary: This study examines how different methods of spatial downscaling of gridded precipitation by climate model impact the projected changes of extreme flood events under different greenhouse gas emission scenarios. We focus on 44 locations within the state of Iowa (central United States) and use a hydrologic model to estimate the changes in flood extremes under current climate and four future emissions scenarios for different downscaling methods. Our results suggest that the choice of downscaling method impacts the magnitude of the projected changes, while the selected emission scenario plays a smaller role on the projected changes. Therefore, we encourage stakeholders to incorporate the impact of downscaling methods within water resource planning and design when considering climate change. Key Points: All emission scenarios and downscaling techniques produce a projected increase in flood extremes across IowaDiscrepancies among downscaling techniques become more pronounced with higher emission scenariosProjected increases in flood extremes become greater as the annual exceedance probability decreases [ABSTRACT FROM AUTHOR]

Design flood estimation is an essential step in many water engineering design tasks such as the planning and design of infrastructure to reduce flood damage. Flood frequency analysis (FFA) is widely used in estimating design floods when the at-site flood data length is adequate. One of the problems in FFA with an annual maxima (AM) modeling approach is deciding how to handle smaller discharge values (outliers) in the selected AM flood series at a given station. The objective of this paper is to explore how the practice of censoring (which involves adjusting for smaller discharge values in FFA) affects flood quantile estimates in FFA. In this regard, two commonly used probability distributions, log-Pearson type 3 (LP3) and generalized extreme value distribution (GEV), are used. The multiple Grubbs and Beck (MGB) test is used to identify low-flow outliers in the selected AM flood series at 582 Australian stream gauging stations. It is found that censoring is required for 71% of the selected stations in using the MGB test with the LP3 distribution. The differences in flood quantile estimates between LP3 (with MGB test and censoring) and GEV distribution (without censoring) increase as the return period reduces. A modest correlation is found (for South Australian catchments) between censoring and the selected catchment characteristics (correlation coefficient: 0.43), with statistically significant associations for the mean annual rainfall and catchment shape factor. The findings of this study will be useful to practicing hydrologists in Australia and other countries to estimate design floods using AM flood data by FFA. Moreover, it may assist in updating Australian Rainfall and Runoff (national guide). [ABSTRACT FROM AUTHOR]

Non-stationary methods of flood frequency analysis are widespread in research but rarely implemented by practitioners. One reason may be that research papers on non-stationary statistical models tend to focus on model fitting rather than extracting the sort of results needed by designers and decision makers. It can be difficult to extract useful results from non-stationary models that include stochastic covariates for which the value in any future year is unknown. We explore the motivation for including such covariates, whether on their own or in addition to a covariate based on time. We set out a method for expressing the results of non-stationary models as an integrated flow estimate, which removes the dependence on the covariates. This can be defined either for a particular year or over a longer period of time. The methods are illustrated by application to a set of 375 river gauges across England and Wales. We find annual rainfall to be a useful covariate at many gauges, sometimes in conjunction with a time-based covariate. For estimating flood frequency in future conditions, we advocate exploring hybrid approaches that combine the best attributes of non-stationary statistical models and simulation models that can represent changes in climate and river catchments. [ABSTRACT FROM AUTHOR]

This paper re-examines the influence of urban development on the number of flood events in Java from 1970 to 2020. This paper aims to prove development policies in Indonesia in the last 50 years and to analyze the correlation between flood events, population growth, and urbanization on a broader regional scale. The Moran-I spatial autocorrelation analysis method is applied to measure its spatial autocorrelation between urbanization and flood events. The dynamic urbanization illustrated by spatial–temporal analysis from the 1970 to the 2020s using Landsat data; in this case, urbanization is measured using the parameters of population concentration and built-up area growth in urban areas of Java. In more detail, the results of hot-spot clusters show development from "lower-low" dispersion in the 70 s to areas with a majority of "higher-high" clustered in 2000–2020. It proves that the development burden in Java is in high category impact, so solutions for population distribution, land development, and spatial planning are needed to ensure environmental sustainability in Java. [ABSTRACT FROM AUTHOR]

The objective of this study was to analyse Manafwa River flood frequency in Eastern Uganda. Analysis of Manafwa River maximum annual flows from 1949-2015 was undertaken using Log Pearson 3 distribution in comparison with Gumbel, Normal and Log Normal distributions to determine frequency of occurrence and magnitude of extreme floods. Statistical analysis including goodness of fit tests of chi-square, Kolmogorov-Smirnov and Anderson-Darling tests were used to generate the most suitable probability distribution model. The results show quantile magnitudes lowest for Log Normal distribution at 43.59 m3/s and highest for Log Pearson 3 distribution at 51.67 m3/s. The 5-year quantile estimates are highest for Normal and Log Pearson at 70.37 m3/s and 63.99 m3/s respectively. The 10-year quantile estimates are highest for Log Normal and lowest for Log Pearson 3 distributions at 87.57 m3/s and 75.13 m3/s respectively. The 100-year quantile estimates are lowest for Normal and highest for Log Normal distributions at 108.57 m3/s and 154.66 m3/s respectively. The 200-year quantile estimates are lowest for Normal and highest for Log Normal distributions respectively at 114.980 m3/s and 177.16 m3/s respectively. Log Pearson 3 distribution emerged as best fit for data. From the statistical analysis, LP 3 probability distribution presents the most accurate regression coefficient at 0.8486 and the most suitable distribution of goodness of best fit using A-D, K-S and Chi square tests followed by the Gumbel distribution. The tests yield 0.15666, 0.04855 and 0.88502 for A-D, K-S and Chi square tests respectively for the LP 3 distribution. There is an increasing upward trend of the discharges at Manafwa River floodplains at higher probabilities of exceedance across all the probability distributions due to varrying climatic changes and rapid landuse changes in the Manafwa catchment. Manafwa river floodplains have the capacity to accommodate and boost crop production and productivity. Any nutrients lost to leaching could be gained from subsequent fallowing and sustainable soil fertility management including; proper drainage, crop rotation, adding organic manure, cover cropping and among others. [ABSTRACT FROM AUTHOR]

Using 4 global reanalysis data sets, significant upward trends of precipitable water vapor (PWV) were found in the 3 time periods of 1958–2020, 1979–2020, and 2000–2020. During 1958–2020, the global PWV trends obtained using the ERA5 and JRA55 data sets are 0.19 ± 0.01 mm per decade (1.15 ± 0.31%) and 0.23 ± 0.01 mm per decade (1.45 ± 0.32%), respectively. The PWV trends obtained using the ERA5, JRA55, NCEP-NCAR, and NCEP-DOE data sets are 0.22 ± 0.01 mm per decade (1.18 ± 0.54%), 0.21 ± 0.00 mm per decade (1.76 ± 0.56%), 0.27 ± 0.01 mm per decade (2.20 ± 0.70%) and 0.28 ± 0.01 mm per decade (2.19 ± 0.70%) for the period 1979–2020. During 2000–2020, the PWV trends obtained using ERA5, JRA55, NCEP-DOE, and NCEP-NCAR data sets are 0.40 ± 0.25 mm per decade (2.66 ± 1.51%), 0.37 ± 0.24 mm per decade (2.19 ± 1.54%), 0.40 ± 0.26 mm per decade (1.96 ± 1.53%) and 0.36 ± 0.25 mm per decade (2.47 ± 1.72%), respectively. Rising PWV has a positive impact on changes in precipitation, increasing the probability of extreme precipitation and then changing the frequency of flood disasters. Therefore, exploring the relationship between PWV (derived from ERA5 and JRA55) change and flood disaster frequency from 1958 to 2020 revealed a significant positive correlation between them, with correlation coefficients of 0.68 and 0.79, respectively, which explains the effect of climate change on the increase in flood disaster frequency to a certain extent. The study can provide a reference for assessing the evolution of flood disasters and predicting their frequency trends.

Abstract This study evaluates five bias correction and statistical downscaling (BCSD) techniques for daily precipitation and examines their impacts on the projected changes in flood extremes (i.e., 1%, 0.5%, and 0.2% floods). We use climate model outputs from the Coupled Model Intercomparison Project Phase 6 (CMIP6) to conduct hydrologic simulations across watersheds in Iowa and determine historical and future flood extreme estimates based on generalized extreme value distribution fitting. Projected changes in these extremes are examined with respect to four Shared Socioeconomic Pathways (SSPs) alongside five BCSD techniques. We find the magnitude of the estimates of future annual exceedance probabilities (AEPs) are expected to increase under all SSPs, especially for the emission scenarios with higher greenhouse gases concentrations (i.e., SSP370 and SSP585). Our results also suggest the choice of BCSD impacts the magnitude of the projected changes, with the SSPs that play a more limited role compared to the choice of downscaling method. The variability in projected flood changes across Iowa is similar across the downscaling technique but increases as the AEP increases. Our findings provide insights into the impact of downscaling techniques on flood extremes' projections and useful information for climate planning across the state.

The correlation between extreme precipitation and flood magnitude is still poorly understood because of the complex mechanisms controlling flood generation. In this study, we used 50 years of precipitation and streamflow data across Iran to analyse the spatial distribution, slope, and significance of long-term trends in extreme precipitation compared to the spatial distribution of long-term trends for flood magnitude and frequency. Despite the decreasing trend of extreme precipitation in 60% of Iran’s subbasins, the flood magnitude showed an increasing trend. In particular, the number of flooding days with a return period of 25~50 years and >50 years in 2010s has increased by 1.3 and 2.2 times, respectively, compared to that in 2000s. We attributed the above contrasting relationship between extreme precipitation and flood magnitude to the influence of an increase in short-duration extreme precipitations, extensive land use and land cover change, and reduced soil infiltration due to long-term droughts. [ABSTRACT FROM AUTHOR]

Estimating peak flow for a catchment is commonly undertaken using the design event method; however, this method does not allow for the understanding of uncertainty in the result. This research first presents a simplified method of fragments approach to rainfall disaggregation that ignores the need to consider seasonality, offering a greater diversity in storm patterns within the resulting sub-daily rainfall. By simulating 20 iterations of the disaggregated sub-daily rainfall within a calibrated continuous simulation hydrologic model, we were able to produce multiple long series of streamflow at the outlet of the catchment. With these data, we investigated the use of both the annual maximum and peaks over threshold approaches to flood frequency analysis and found that for a 1-in-100-year annual exceedance probability peak flow, the peaks over threshold method (333 m3/s ± 50 m3/s) was significantly less uncertain than the annual maximum method (427 m3/s ± 100 m3/s). For the 1-in-100-year annual exceedance probability, the median peak flow from the peaks over threshold method (333 m3/s) produced an outcome comparable to the design event method peak flow (328 m3/s), indicating that this research offers an alternative approach to estimating peak flow, with the additional benefit of understanding the uncertainty in the estimation. Finally, this paper highlighted the impact that length and period of streamflow has on peak flow estimation and noted that previous assumptions around the minimum length of gauged streamflow required for flood frequency analysis may not be appropriate in particular catchments. [ABSTRACT FROM AUTHOR]

Abstract Flooding is recognized worldwide joined of the most expensive natural hazards. To adopt proper structural and nonstructural measurements for controlling and mitigating the rising flood risk, the availability of streamflow values along a river is essential. This raises concerns in the hydrological assessment of poorly gauged or ungauged catchments. In this regard, several flood frequency analysis approaches have been conducted in the literature including index flow method (IFM), square grids method (SGM), hybrid method (HM), as well as the conventional multivariate regression method (MRM). While these approaches are often based on assumptions that simplify the complex nature of the hydrological system, they might not be able to address uncertainties associated with the complexity of the system. One of the powerful tools to deal with this issue is data-driven model that can be easily adopted in complex systems. The objective of this research is to utilize three different data-driven models: random forest (RF), adaptive neuro-fuzzy inference system (ANFIS), and M5 decision tree algorithm to predict peak flow associated with various return periods in ungauged catchments. Results from each data-driven model were assessed and compared with the conventional multivariate regression method. Results revealed all the three data-driven models performed better than the multivariate regression method. Among them, the RF model not only demonstrated the superior performance of peak flow prediction compared to the other algorithms but also provided insight into the complexity of the system through delivering a mathematical formulation.

This present study investigates different techniques for estimating the surface runoff using the Soil Conservation Service Curve Number (SCS-CN) method for the Aghanashini River in Karnataka, India. The SCS-CN method is a simplified approach for runoff estimation, but it does not take into account the actual moisture content in the soil. Consequently, insignificant moisture level changes could induce significant variations in the runoff. The study analyzes six different models based on the SCS-CN method, including the original SCS-CN model and several variations with added features (SCS-CN with slope correction, SCS-CN with λ-optimization, Mishra and Singh, Michel-Vazken -Perrin (MVP), Activation Soil Moisture Accounting SCS-CN). The accuracy of each model was compared using several goodness-of-fit statistics. Furthermore, based on the flood frequency analysis, three large flood events were reported in 2005, 2013, and 2014. The results showed that the MVP model was the best-performing method in simulating runoff. The outcomes of this study can provide valuable information to the local authorities in making informed decisions about flood forecasting and water conservation. HIGHLIGHTS Six mathematical models have been prepared on the basis of SCS-CN for a coastal river basin.; The long-term hydrological simulation of the Aghanashini River has been carried out by taking AMC changes.; Seven statistical indices were used to judge the efficiency of the developed models.; The developed models compute surface runoff with the desired accuracy.;

This study explores causal mechanisms of river metamorphosis and its impacts on regional landscapes. The study also investigates the implications of metamorphosis on associated ecological resources. Advanced GIS and remote sensing technologies were used to delineate morphological parameters describing metamorphosis of the Old Brahmaputra River from historical maps (i.e., Rannell's Map in 1776, Tassin's Map in 1840, Topographic Survey Map in 1943) and remotely sensed optical satellite imagery Sentinel-2 in 2022. Flood frequencies were investigated for different periods by applying Gumbel's Analytical Method (GAM), Log-Pearson Type III, and Log-Normal Method to estimate probability of flood vulnerability and impacts of flooding on morphodynamics in the central Bengal Basin. During the periods between 1776 and 2022, the area of sedimentation (77,999.43 ha) was greater than the eroded area (2983.29 ha).This difference was attributed to siltation of the channel bed morphology and corresponding accelerated flood vulnerability that accompanied river metamorphosis. Hydrological variables particularly annual average discharge significantly declined from 22 to ~ 18 m3/s per year during the period from 1965 to 2020. The study results demonstrated that the log-normal methods significantly overestimated peak flood discharge compared to Log-Pearson methods and Gumbel's probability model. The extrapolation of the discharge for the 100-year flood by applying the three methods produced values of 712.66 m3/s, 1750.26 m3/s, and 2462.92 m3/s. Differences of these magnitudes may be critical for planning purposes because these differences in results will generate large-scale projected impacts on morphodynamics of the central Bengal Basin. [ABSTRACT FROM AUTHOR]

Despite calls for large‐scale afforestation to alleviate flooding, the effectiveness of such action remains unclear. Simulations with the SHETRAN hydrological model are therefore carried out for the 335‐km2 Irthing catchment and its 1‐10‐km2 headwater catchments in northwest England to determine: (a) whether forests can reduce flood peak discharges in a large catchment; (b) the proportion of the catchment that requires afforestation to be effective; and (c) the extent to which a combination of afforestation and natural flood management features (leaky dams) improves upon afforestation on its own. Four‐year simulations were run with a range of forest covers and extents of leaky dam installation, the latter modelled as a channel hydraulic resistance. Hydrograph, flood frequency and peak discharge magnitude responses to forest cover simulated (and observed) in the headwater catchments are replicated in simulations at the full scale. Afforestation on its own can reduce the frequency of given flood magnitudes but has a variable impact on individual peak discharge magnitudes. For the Irthing, a 76% forest cover reduces the mean discharge of 20 peaks in a partial duration flood series by 10% relative to the current 21% forest cover but reduces the largest peak by only 2.5%. Accompanying adverse effects include 17.5% reduction in long‐term runoff and loss of agricultural land. By contrast, leaky dams mitigate flood frequencies and peak discharges effectively, over a range of discharge magnitudes, with no reduction of annual runoff. The dam installation required for a 10% mean peak discharge reduction reduces the largest peak by 22%. The flow resistance increases by which the dams are simulated have still to be translated into specific dam designs. Nevertheless, and considering the figures indicatively, dam installation in 20% of the Irthing streams with Strahler orders of 1–3 achieves a 10% peak discharge reduction with only 35% forest cover and 7.5% runoff reduction. The study illustrates the potential for a dense network of leaky dams to augment the impact of afforestation on flood mitigation, especially at the largest discharges, while minimizing adverse impacts on water resources and food security. [ABSTRACT FROM AUTHOR]

Flooding is recognized worldwide joined of the most expensive natural hazards. To adopt proper structural and nonstructural measurements for controlling and mitigating the rising flood risk, the availability of streamflow values along a river is essential. This raises concerns in the hydrological assessment of poorly gauged or ungauged catchments. In this regard, several flood frequency analysis approaches have been conducted in the literature including index flow method (IFM), square grids method (SGM), hybrid method (HM), as well as the conventional multivariate regression method (MRM). While these approaches are often based on assumptions that simplify the complex nature of the hydrological system, they might not be able to address uncertainties associated with the complexity of the system. One of the powerful tools to deal with this issue is data-driven model that can be easily adopted in complex systems. The objective of this research is to utilize three different data-driven models: random forest (RF), adaptive neuro-fuzzy inference system (ANFIS), and M5 decision tree algorithm to predict peak flow associated with various return periods in ungauged catchments. Results from each data-driven model were assessed and compared with the conventional multivariate regression method. Results revealed all the three data-driven models performed better than the multivariate regression method. Among them, the RF model not only demonstrated the superior performance of peak flow prediction compared to the other algorithms but also provided insight into the complexity of the system through delivering a mathematical formulation. [ABSTRACT FROM AUTHOR]

Accurate modeling of infiltration losses is vital for runoff estimation and thus the development of flood design/protection criteria and water management schemes, etc. In design flood practices, the initial abstraction and constant loss (IACL) method has been widely applied due to its simplicity. However, due to a lack of physical equivalent properties, the IACL method is often subject to issues in parametrization and has large dependency on calibration efforts for storm events. Despite the wide range/variability of IACL values, a single set of IA and CL values is normally adopted for specific flood frequency, which may introduce uncertainty and bias in resulting peak streamflow. In this study, we identified a total of 2,036 rainfall-runoff events for 18 watersheds in North Central Texas to estimate the total losses with their IA and CL components based on time-series of mean areal precipitation (MAP) and streamflow data. Threshold behavior is found for all studied subbasins between the summation of gross rainfall and antecedent soil moisture versus runoff depth: below the threshold, runoff depth is minimal; whereas above it, runoff is largely linearly correlated with the summation of rainfall and antecedent soil moisture. This finding provides a convenient way to estimate/predict total loss or runoff depth given MAP and antecedent soil moisture. In addition, this study shows that the IA and CL values can be approximated by the gamma and Weibull distributions, respectively. The fitted distributions of IA and CL values can be applied in a Monte Carlo simulation framework to stochastically simulate numerous rainfall-runoff events for a flood frequency analysis. [ABSTRACT FROM AUTHOR]

Hydrology and meteorological data from relatively undisturbed watersheds aid in identifying effects on ecosystem services, tracking hydroclimatic trends, and reducing model uncertainties. Sustainable forest, water, and infrastructure management depends on assessing the impacts of extreme events and land use change on flooding, droughts, and biogeochemical processes. For example, global climate models predict more frequent high-intensity storms and longer dry periods for the southeastern USA. We summarized 17 years (2005–2021) of hydrometeorological data recorded in the 52 km2, third-order Turkey Creek watershed at the Santee Experimental Forest (SEF), Southeastern Coastal Plain, USA. This is a non-tidal headwater system of the Charleston Harbor estuary. The study period included a wide range of weather conditions; annual precipitation (P) and potential evapotranspiration (PET) ranged from 994 mm and 1212 mm in 2007 to 2243 mm and 1063 in 2015, respectively. The annual runoff coefficient (ROC) varied from 0.09 in 2007 (with water table (WT) as deep as 2.4 m below surface) to 0.52 in 2015 (with frequently ponded WT conditions), with an average of 0.22. Although the average P (1470 mm) was 11% higher than the historic 1964–1976 average (1320 mm), no significant (α= 0.05) trend was found in the annual P (p = 0.11), ROC (p = 0.17) or runoff (p = 0.27). Runoff occurred on 76.4% of all days in the study period, exceeding 20 mm/day for 1.25% of all days, mostly due to intense storms in the summer and lower ET demand in the winter. No-flow conditions were common during most of the summer growing season. WT recharge occurred during water-surplus conditions, and storm-event base flow contributed 23–47% of the total runoff as estimated using a hydrograph separation method. Storm-event peak discharge in the Turkey Creek was dominated by shallow subsurface runoff and was correlated with 48 h precipitation totals. Estimated precipitation intensity–duration–frequency and flood frequency relationships were found to be larger than those found by NOAA for the 1893–2002 period (for durations ≥ 3 h), and by USGS regional frequencies (for ≥10-year return intervals), respectively, for the same location. We recommend an integrated analysis of these data together with available water quality data to (1) assess the impacts of rising tides on the hydroperiod and biogeochemical processes in riparian forests of the estuary headwaters, (2) validate rainfall–runoff models including watershed scale models to assess land use and climate change on hydrology and water quality, and (3) inform watershed restoration goals, strategies, and infrastructure design in coastal watersheds.

The flood regimes have been changing due to the climate change and human activities. Evaluating the flood risk under nonstationarity is critical to water resource management authorities in disaster reduction. In this study, the annual maximum streamflow (AMS) was used to analyze the nonstationarity in flood frequency in the Hanjiang River (HJR) Basin. A Generalized Additive Model for Location, Scale and Shape (GAMLSS) is employed to characterize nonstationarity in the AMS with time, as well as climatic and anthropogenic factors. Additionally, changes of the AMS in response to future climate change and human activity are also investigated. Results indicate that flood behavior can be better described by the nonstationary model with physically based covariates than that with time and the stationary model, implying that flood regimes of the HJR Basin are mainly influenced by anthropogenic and climatic factors including reservoir, precipitation and temperature. The precipitation and temperature projected by the Statistical Downscaling Model (SDSM) under two climate scenarios (RCP2.6 and RCP4.5) in the HJR Basin are characterized by an increasing trend over the period of 2006–2100. Furthermore, an increasing trend was found in the AMS during 2051–2100, indicating that flood risk is likely to increase in the future in the HJR Basin due to the climate change alone without further changes in hydrological engineering and flood management. The results quantified the flood frequency under nonstationarity conditions with physically based covariates and provided information to the decision-makers to address the potential risks posed to the HJR Basin. HIGHLIGHTS The nonstationary model with physically based explanatory covariates can better describe the flood regime.; The model analyzes the changes in the annual maximum streamflow over time and association with climatic and reservoir indices.; Flood risk is likely to increase in the future in the Hanjiang River Basin due to climate change alone without further changes in hydrological engineering and flood management.;

Climate change models for southern Africa predict less frequent, but more intense, rainfall events, and an increased frequency of tropical cyclones. With their steep topography and small catchments, headwater streams generate large floods following intense rainfall events. Large flooding events in headwater streams are under studied in southern Africa. In this paper, we explore flooding in the upper Klaserie River, Limpopo River System, South Africa to determine the flow distribution and flood frequency for the catchment. In addition, we determine the return level for a large, economically damaging, flood generated following the landfall of a sub-tropical depression in January 2012 and, attempt to identify rainfall patterns that resulted in similar floods. An annual hydrological cycle with summer maxima and winter minima for both rainfall and flow was identified. The flood frequency analysis demonstrated that the January 2012 flood had an estimated return level of 225 years. This flood had a peak flowrate exceeding 1200 m3s-1 in a system with an average daily flowrate of 1 m3s-1. Regression tree analysis showed that a two-day rainfall in excess of 240 was a predictor for four of the five largest floods. A two-day rainfall in excess of 400 mm distinguished the January 2012 flood from other floods. Non-stationarity analyses for the flow and rainfall data and a SWAT hydrological model are recommend for the upper Klaserie River to evaluate climate and land cover changes, and their relationship to the magnitude of the 2012 flood. Our study demonstrates that South African river monitoring data can be used to detect and characterize major floods, despite deficiencies in these data. Continuation of these monitoring programs is vital for river health monitoring and the detection of trends in floods resulting from human activities and climate change.

We present a timeseries of flood and slumping phases in central Europe for the past 65,000 years from event layers in sediment cores from infilled Eifel maar basins (Germany). Palynological, petrographic and organic carbon (chlorins) records are used to understand the precise timing of these events. Periods of increased flood activity seem to coincide with Heinrich stadials in marine sediment records, which are associated with cold and more arid climate conditions, indicating a vegetation response within the maars' catchment areas. This multi-proxy correlation reveals prominent slumps at different maar sites during Greenland Stadial (GS) 12. The stratigraphy is based on sediment records from the Auel infilled maar and we thus call this event Auel Cold Event (ACE). Frozen and fractured sediment packages within the slump suggest deep frost or permafrost conditions for the region during the stadial. The results agree well with sediment archives and archeological sites across Europe that report severely cold and arid conditions for the stadial. This supports the assumption that GS12 was indeed one of the coldest periods of the last glacial cycle rather than the Heinrich stadials. Based on our age model, the ACE occurred at 43,500 yr b2k (years before the year 2000), which coincides with the initial weakening of Earth's magnetic field strength prior to the Laschamp geomagnetic excursion. [ABSTRACT FROM AUTHOR]

Abstract. A reliable flood frequency analysis (FFA) requires selection of an appropriate statistical distribution to model historic streamflow data and, where streamflow data are not available (ungauged sites), a regression-based regional flood frequency analysis (RFFA) often correlates well with downstream channel discharge to drainage area relations. However, the predictive strength of the accepted RFFA relies on an assumption of homogeneous watershed conditions. For glacially conditioned fluvial systems, inherited glacial landforms, sediments, and variable land use can alter flow paths and modify flow regimes. This study compares a multi-variate RFFA that considers 28 explanatory variables to characterize variable watershed conditions (i.e., surficial geology, climate, topography, and land use) to an accepted power-law relationship between discharge and drainage area. Archived gauge data from southern Ontario, Canada are used to test these ideas. Mathematical goodness of-fit criteria best estimate flood discharge for a broad range of flood recurrence intervals, i.e., 1.25, 2, 5, 10, 25, 50, and 100 years. The LN, EV1, LP3, and GEV distributions are found most appropriate in 42.5%, 31.9%, 21.7%, and 3.9% of cases, respectively, suggesting that systematic model selection criterion is required for FFA in heterogeneous landscapes. Multivariate regression of estimated flood quantiles with backward elimination of explanatory variables using principal component and discriminant analyses reveal that precipitation provides a greater predictive relationship for more frequent flood events, whereas surficial geology demonstrates more predictive ability for high magnitude, less frequent flood events. In this study, all seven flood quantiles identify a statistically significant two-predictor model that incorporates upstream drainage area and the percentage of naturalized landscape with 5% improvement in predictive power over the commonly used single-variable drainage area model (p < 2.2e-16). An analysis of variance (ANOVA) further supports the two-predictor model indicating a decrease in the sum of squares of residuals and an F statistic (p < 0.001) that demonstrates very strong evidence in favour of the two-predictor model (i.e., drainage area and land use) when estimating flood discharge in this low-relief landscape with pronounced glacial legacy effects and heterogenous land use. [ABSTRACT FROM AUTHOR]

For practical purposes the current guidance for attributing fluvial climate change allowances (recommended peak-discharge adjustments) in England is based on percentage changes to the 50-year return level attributed homogeneously to geographical regions. This proof-of-concept study introduces an approach enabling the practitioner to derive allowances for the full distribution of extremes (as opposed to single return periods) and is based on catchment characteristics. A region of influence approach, adopted by the Flood Estimation Handbook (FEH), is applied to estimate changes to the parameters of statistical models of extreme flows, as opposed to flow peaks. The approach is distribution neutral and can be applied to any catchment for which an FEH analysis has been undertaken. Results for an example scenario (Representative Concentration Pathway (RCP)8.5 2080s) are compared to a geographically regional method and an example is provided for a single catchment estimating extremes with the FEH method and adjusting them for the RCP8.5 2080s scenario. [ABSTRACT FROM AUTHOR]

Frequency analysis of extreme events is used to evaluate the probable maximum flood from runoff records using probability distribution functions. This study aims to understand rainfall-runoff relationships in four conterminous basins in Southwestern Nigeria using the runs test, Pearson correlation, and wavelet coherence. The study revealed that the highest Extreme Annual Rainfall in the series is expected to occur once every 31 years with 0.03 probability. Wavelet coherence shows a strong significant association with phases between Yewa – Ikeja and Yewa – Abeokuta. The observed extreme discharge values constitute the basis for establishing exceedance probabilities for the stations and their recurrence intervals. [ABSTRACT FROM AUTHOR]

The estimation of reservoirs impact on flood peak reduction at the catchment scale is of fundamental importance for risk assessment and planning purposes. It is generally addressed using detailed hydrologic‐hydrodynamic simulations or simple, empirically based indices. The former provide detailed results, at the cost of a large amount of information and high computational efforts; the latter are cost‐effective and simple, yet they generally provide approximate results. A promising compromise between the two is the physically based attenuation index R $\mathcal{R}$, based on the concept of equivalent reservoir; R $\mathcal{R}$ was recently proposed in the literature to estimate the impact of multiple reservoirs located in series along the main channel, based on the assumption of rectangular catchment. In this work, we extend the equivalent reservoir approach to a generic catchment, with any shape and without restrictions on the location of the reservoirs. For an effective assessment of the method, we also introduce the novel concept of Reservoir‐influenced Instantaneous Unit Hydrograph (RIUH); the RIUH can be derived using a Monte Carlo procedure to include the effects of reservoirs on the Instantaneous Unit Hydrograph. Simulations of numerous fictitious reservoir configurations in a real basin demonstrate the potential of the methods in reproducing the attenuation effect. Remarkably, an average error of 3%–5% exists between R $\mathcal{R}$ and the peak reduction of a less simplified reality (RIUH). Finally, we provide a global, catchment‐scale application of the attenuation index, which constitutes an effective tool for the evaluation of each reservoir in managed catchment system and the design or planning of new hydraulic infrastructures. Key Points: A simplified reservoir attenuation index R $\mathcal{R}$, applicable to a generic catchment and any reservoir location, is proposedWe introduce the novel concept of Reservoir‐influenced Instantaneous Unit Hydrograph to easily include reservoirs effects in Instantaneous Unit HydrographWe provide a global, catchment‐scale application of R $\mathcal{R}$ that constitutes an effective tool for risk assessment and reservoir design [ABSTRACT FROM AUTHOR]

Climate change models for southern Africa predict less frequent, but more intense, rainfall events, and an increased frequency of tropical cyclones. With their steep topography and small catchments, headwater streams generate large floods following intense rainfall events. Large flooding events in headwater streams are under studied in southern Africa. In this paper, we explore flooding in the upper Klaserie River, Limpopo River System, South Africa, to determine the flow distribution and flood frequency for the catchment. In addition, we determine the return level for a large, economically damaging, flood generated following the landfall of a sub-tropical depression in January 2012 and, attempt to identify rainfall patterns that resulted in similar floods. An annual hydrological cycle with summer maxima and winter minima for both rainfall and flow was identified. The flood frequency analysis demonstrated that the January 2012 flood had an estimated return level of 225 years. This flood had a peak flowrate exceeding 1200 m3s-1 in a system with an average daily flowrate of 1 m3s-1. Regression tree analysis showed that a two-day rainfall in excess of 240 was a predictor for four of the five largest floods. A two-day rainfall in excess of 400 mm distinguished the January 2012 flood from other floods. Non-stationarity analyses for the flow and rainfall data and a SWAT hydrological model are recommend for the upper Klaserie River to evaluate climate and land cover changes, and their relationship to the magnitude of the 2012 flood. Our study demonstrates that South African river monitoring data can be used to detect and characterize major floods, despite deficiencies in these data. Continuation of these monitoring programs is vital for river health monitoring and the detection of trends in floods resulting from human activities and climate change. [ABSTRACT FROM AUTHOR]

Reducing uncertainty in design flood estimates is an essential part of flood risk planning and management. This study presents results from flood frequency estimates and associated uncertainties for five commonly used probability distribution functions, extreme value type 1 (EV1), generalized extreme value (GEV), generalized pareto distribution (GPD), log normal (LN) and log Pearson type 3 (LP3). The study was conducted using Monte Carlo simulation (MCS) and bootstrapping (BS) methods for the 10 river catchments in eastern Australia. The parameters were estimated by applying the method of moments (for LP3, LN, and EV1) and L-moments (for GEV and GPD). Three-parameter distributions (e.g., LP3, GEV, and GPD) demonstrate a consistent estimation of confidence interval (CI), whereas two-parameter distributions show biased estimation. The results of this study also highlight the difficulty in flood frequency analysis, e.g., different probability distributions perform quite differently even in a smaller geographical area. [ABSTRACT FROM AUTHOR]

Floods are one of the world's most destructive natural disasters, taking more lives and causing more infrastructural damage than any other natural phenomenon. Floods have a significant economic, social, and environmental impact in developing countries like India. As a result, it is essential to address this natural disaster to mitigate its effects. The lower Narmada basin has experienced numerous floods, including severe flooding in 1970, 1973, 1984, 1990, 1994, and 2013. The objective of the present study is to use flood frequency analysis to anticipate peak floods and prepare flood inundation maps for the lower Narmada River reach. The flood frequency analysis was carried out using Gumbel's and Log-Pearson Type III Distribution methods. The hydrodynamic simulation was performed using HEC-RAS v6.0 to prepare flood inundation maps for predicted flood peaks. The result shows that the Log-Pearson Type-III distribution method gives good results for the lower return period while Gumbel's method gives good results for the higher return period. The hydrodynamic model results indicate that as the return period increases, the area of the high-risk zone increases while the area of the low-risk zone remains almost constant. The present study concludes that the existing embankment system on the banks of the Narmada River is not sufficient for significant floods. The developed maps will be helpful to government authorities and individual stakeholders to decide the flood mitigation measures. HIGHLIGHTS The present study focuses on identifying the impact of the flood in the Lower Narmada Basin, India.; The statistical analysis such as Gumbel and Log-Pearson methods were utilized for flood frequency analysis.; The hydrodynamic simulation was carried out using HEC-RAS for inundation mapping and identifying flood risk areas.; The results from the present study will help decide mitigation measures in the region.;

The UK standard for estimating flood frequencies is outlined by the flood estimation handbook (FEH) and associated updates. Estimates inevitably come with uncertainty due to sampling error as well as model and measurement error. Using resampling approaches adapted to the FEH methods, this paper quantifies the sampling uncertainty for single site, pooled (ungauged), enhanced single site (gauged pooling) and across catchment types. This study builds upon previous progress regarding easily applicable quantifications of FEH-based uncertainty estimation. Where these previous studies have provided simple analytical expressions for quantifying uncertainty for single site and ungauged design flow estimates, this study provides an easy-to-use method for quantifying uncertainty for enhanced single site estimates. HIGHLIGHTS Bespoke bootstrap methods for quantifying uncertainty for ungauged and enhanced single site FEH design flow estimation.; Comparison of flood estimation uncertainty across catchment types.; Simple equations to derive variance and standard error for enhanced single site design flow estimates.;

Over the past half-century, the risk of urban flooding in Dar es Salaam has increased due to changes in land cover coupled with climatic changes. This paper aimed to quantify the impacts of climate and land-cover changes on the magnitudes and frequencies of flood runoffs in urban Dar es Salaam, Tanzania. A calibrated and validated SWAT rainfall-runoff model was used to generate flood hydrographs for the period 1969–2050 using historical rainfall data and projected rainfall based on the CORDEX-Africa regional climate model. Results showed that climate change has a greater impact on change in peak flows than land-cover change when the two are treated separately in theory. It was observed that, in the past, the probability of occurrence of urban flooding in the study area was likely to be increased up to 1.5-fold by climate change relative to land-cover change. In the future, this figure is estimated to decrease to 1.1-fold. The coupled effects of climate and land-cover changes cause a much bigger impact on change in peak flows than any separate scenario; this scenario represents the actual scenario on the ground. From the combined effects of climate and land-cover changes, the magnitudes of mean peak flows were determined to increase between 34.4 and 58.6% in the future relative to the past. However, the change in peak flows from combined effects of climate and land-cover changes will decrease by 36.3% in the future relative to the past; owing to the lesser variations in climate and land-cover changes in the future compared with those of the past. HIGHLIGHTS Investigate temporal variability of urban flood runoffs from the impacts of climate and land-cover changes.; Compare past and future peak flow trend magnitudes based on CORDEX-Africa RCM under RCP4.5.; Investigate changes in the historical and future intensities of urban flood runoffs.; Illustrate the probabilistic impacts of climate and land-cover changes on the recurrence intervals of urban floods.;

• The regulation of upstream cascade reservoirs directly affects downstream floods. • We propose a modified reservoir index. • We propose two flood frequency analysis methods considering flood staging. Reservoirs alter the flood propagation process. Considering that many reservoirs have different flood limit water levels and flood control storage capacities during different flood seasons, we propose a modified reservoir index. In this study, we construct four types of non-stationary statistical models, taking time t, antecedent rainfall P, and the modified Reservoir Index MRI as candidate explanatory variables. Then we compare the simulation effects of different models. The equivalent reliability method and the extreme value distribution method considering the flood staging are proposed to calculate the flood design values at different return periods for downstream sites of the reservoir. The calculation results at the Huangzhuang Hydrological Station in the lower Hanjiang River in China show that the flood occurrence time in the upper Hanjiang River and its corresponding reservoir seasonal flood limit level and flood control storage capacity directly affect the flood in the downstream river. The non-stationary model incorporating P and MRI as explanatory variables achieves the best fit for the flood. Compared with the existing reservoir indices, MRI can better describe the impact of the upstream reservoir group on the flood sequence. The calculation results of the equivalent reliability method and the extreme value distribution method are comparable. The impact of the staged setting of flood limit water levels should be fully considered. The results of the equal reliability method and the extreme value distribution method can provide a reference for the design of flood control projects. [ABSTRACT FROM AUTHOR]