Gauging extreme floods on YouTube: application of LSPIV to home movies for the post‐event determination of stream discharges

Movies taken by witnesses of extreme flood events are increasingly available on video sharing websites. They potentially provide highly valuable information on flow velocities and hydraulic processes that can help improve the post‐flood determination of discharges in streams and flooded areas. We investigated the troubles and potential of applying the now mature large‐scale particle image velocimetry (LSPIV) technique to such flood movies that are recorded under non‐ideal conditions. Processing was performed using user‐friendly, free software only, such as Fudaa‐LSPIV. Typical issues related to the image processing and to the hydrological analysis are illustrated using a selected example of a pulsed flash‐flood flow filmed in a mountainous torrent. Simple corrections for lens distortion (fisheye) and limited incoherent camera movement (shake) were successfully applied, and the related errors were reduced to a few percents. Testing the different image resolution levels offered by YouTube showed that the difference in time‐averaged longitudinal velocity was less than 5% compared with full resolution. A limited number of GRPs, typically 10, is required, but they must be adequately distributed around the area of interest. The indirect determination of the water level is the main source of uncertainty in the results, usually much more than errors because of the longitudinal slope and waviness of the free‐surface of the flow. The image‐based method yielded direct discharge estimates of the base flow between pulses, of the pulse waves, and of the time‐averaged flow over a movie sequence including a series of five pulses. A comparison with traditional indirect determination methods showed that the critical‐depth method may produce significantly biassed results for such a fast, unsteady flow, while the slope‐area method seems to be more robust but would overestimate the time‐averaged flow rate if applied to the high‐water marks of a pulsed flow. Copyright © 2015 John Wiley & Sons, Ltd.     

Fully coupled free‐surface flow and sediment transport modelling of flash floods in a desert stream in the Mojave Desert,California

For most of the year, a dry‐bed desert wash is void of water flow. Intensive rain events, however, could trigger significant flash floods that bring about highly complicated hydrodynamics and morphodynamics processes within a desert stream. We present a fully coupled three‐phase flow model of air, water, and sediment to simulate numerically the propagation of a flash flood in a field‐scale fluvial desert stream, the so‐called Tex Wash located in the Mojave Desert, California, United States. The turbulent flow of the flash flood is computed using the three‐dimensional unsteady Reynolds‐averaged Navier–Stokes equations closed with the shear stress transport k ? ω model. The free surface of the flash flood at the interface of air and water phases is computed with the level‐set method, which enables instantaneous tracking of the water surface as the flash flood propagates over the dry bed of the desert stream. The evolution of the desert fluvial stream's morphology, due to the action of the propagating flash flood on the mobile bed, is calculated using a Eulerian morphodynamics model based on the curvilinear immersed boundary method. The capabilities of the proposed numerical framework are demonstrated by applying it to simulate a flash flood event in a 0.65‐km ‐long reach of the Tex Wash, the intricate channel morphology of which is obtained using light imaging detection and ranging technology. The simulated region of the stream includes a number of bridge foundations. The simulation results of the model for the flash flood event revealed the formation of a highly complex flow field and scour patterns within the stream. Moreover, our simulation results showed that most scour processes take place during the steady phase of the flash flood, that is, after the flash flood fills the stream. The transient phase of the flash flood is rather short and contributes to a very limited amount of erosion within the desert stream.     

Real‐time monitoring and estimation of the discharge of flash floods in a steep mountain catchment

Synchronously and accurately estimating the flood discharges and dynamic changes in the fluid density is essential for hydraulic analysis and forecasting of flash floods, as well as for risk assessment. However, such information is rare for steep mountain catchments, especially in regions that are hotspots for earthquakes. Therefore, six hydrological monitoring sites were established in the main stream and tributaries of the 78.3‐km² Longxi River catchment, an affected region of the Wenchuan earthquake region in China. Direct real‐time monitoring equipment was installed to measure the flow depths, velocities, and fluid total pressures of the flood hydrographs. On the basis of field measurements, real‐time mean cross‐sectional velocities during the flood hydrographs could be derived from easily obtainable parameters: cross‐sectional maximum velocities and the calibrated dimensionless parameter K_h . Real‐time discharges were determined on the basis of a noncontact method to establish the effective rating curves of this mountainous stream, ranging from 1.46 to 386.34 m³/s with the root mean square errors of ≤10.22 m³/s. Compared with the traditional point‐velocity method and empirical Manning's formula, the proposed noncontact method was reliable and safe for monitoring whole flood hydrographs. Additionally, the real‐time fluid density during the flood hydrographs was calculated on the basis of the direct monitoring parameters for fluid total pressures and water depths. During the flood hydrograph, transient flow behaviour with higher fluid density generally occurred downstream during the flood peak periods when the flow was in the supercritical flow regime. The observed behaviour greatly increased the threat of damage to infrastructure and human life near the river. Thus, it is important to accurately estimate flood discharge and identify for fluid densities so that people at risk from an impending flash flood are given reliable, advanced warning.     

Statistical analysis of flash flood events for designing water harvesting systems in an extremely arid environment

A water harvesting system for research purposes has been established in the Lisan Peninsula of the Dead Sea in the middle of the Jordan Rift Valley, where no authorized guideline is available for designing water harvesting systems. Rainfall and runoff, which occurred as flash floods, were observed at the downstream end of a gorge with a 1.12 km² barren catchment area from October 2014 through July 2019. Due to the extremely arid environment, runoff from the catchment is ephemeral, and the flash flood events can be clearly distinguishable from each other. Thirteen flash flood events with a total runoff volume of more than 100 m³ were successfully recorded during the five rainy seasons. Pearson and Spearman correlations between duration, total rainfall depths at two points, total runoff volume, maximum runoff discharge, bulk runoff coefficient, total variation in runoff discharge and maximum variation in runoff discharge of each flash flood event were examined, revealing no straightforward relationship between rainfall and runoff. The performance of the conventional SCS runoff curve number method was also deficient in reproducing any rainfall–runoff relationship. Therefore, probability distribution fitting was performed for each random variable, focusing on the lognormal distribution with three parameters and the generalized extreme value distribution. The maximum goodness-of-fit estimation turns out to be a more rational and efficient method in obtaining the parameter values of those probability distributions rather than the standard maximum likelihood estimation, which has known disadvantages. Results support the design of the water harvesting system and provide quantitative information for designing and operating similar systems in the future.     

Large-scale particle tracking velocimetry with multi-channel CCD cameras

This paper presents a large-scale particle tracking velocimetry(LSPTV) system for measuring surface velocity in vast unsteady flows.The system consists of a flexible number of one-computer-six- camera working units connected via local network or the internet.Multiple zero-tilting cameras are used to increases the field of view with minimum perspective distortions.External synchronization of video outputs of all cameras facilitates uninterrupted image acquisition within the system.Effectiveness and efficiency of the multi-channel LSPTV are enhanced by optimizing illumination,flow seeding, image acquisition and correction,tracer identification,particle matching,and post-processing.The capacity of the LSPTV was demonstrated in the measurement of physical model flows for the Three Gorges Project.     

Estimating the uncertainty of video-based flow velocity and discharge measurements due to the conversion of field to image coordinates

Video-based hydrometry continues to develop for contactless discharge measurements, automated flood gauging stations and the use of crowd-sourced flood videos for discharge reconstruction. Irrespective of the velocimetry algorithm used (LSPIV, STIV, PTV…), orthorectification of the images is necessary beforehand, so that each pixel has the same known physical size. Most times, the orthorectification transformation is a plane-to-plane projection from the water surface to the camera sensor. Two approaches are typically used to compute the coefficients of this transformation: their calibration from ground reference points (GRPs) with known image and real-world coordinates (“implicit calibration”) or their calculation from the values of the intrinsic (focal length, sensor size) and extrinsic (position, angles) parameters of the camera (“explicit calibration”). In this paper, we develop a Bayesian method which makes it possible to combine the implicit and explicit approaches in a probabilistic framework. The Bayesian approach can be used from situations suitable for the implicit approach (plenty of GRPs) to situations propitious to the explicit approach (well-known camera parameters). The method is illustrated using synthetic views of a typical streamgauging scene with known true values of the parameters and GRP coordinates. We show that combining observational and prior information is generally beneficial to get precise estimates. Further tests carried out with a real scene of the Arc River at Randens, France, in flood conditions illustrate the impact of the number, uncertainty and spatial distribution of GRPs on the final uncertainty of flow velocity and discharge.     

Cross-correlation of random fields: mathematical approach and applications

Random field cross‐correlation is a new promising technique for seismic exploration, as it bypasses shortcomings of usual active methods. Seismic noise can be considered as a reproducible, stationary in time, natural source. In the present paper we show why and how cross‐correlation of noise records can be used for geophysical imaging. We discuss the theoretical conditions required to observe the emergence of the Green's functions between two receivers from the cross‐correlation of noise records. We present examples of seismic imaging using reconstructed surface waves from regional to local scales. We also show an application using body waves extracted from records of a small‐scale network. We then introduce a new way to achieve surface wave seismic experiments using cross‐correlation of unsynchronized sources. At a laboratory scale, we demonstrate that body wave extraction may also be used to image buried scatterers. These works show the feasibility of passive imaging from noise cross‐correlation at different scales.     

A simple image‐based strain measurement method for measuring the strain fields in an RC‐wall experiment

A simple image‐based method for measuring plane strain fields on the surface of specimens in earthquake engineering experiments was developed. This method integrated camera calibration, stereo triangulation, image metric rectification and image template matching techniques to develop a method that was cost‐effective, easy to apply and provided a satisfactory level of measurement accuracy. A zero‐strain test conducted using this method showed that the measurement accuracy achieved was 0.04 pixels. That is, the relative displacement accuracy achieved was 0.005 mm and the strain accuracy was 0.001. This level of accuracy was achieved using eight‐mega‐pixel digital cameras to measure a 17 cm × 28 cm measurement region. Cracks that were 0.012 mm wide were identified in the concrete by examining the displacement fields calculated through the application of this image‐based method in an RC‐wall experiment. Copyright © 2011 John Wiley & Sons, Ltd.     

Radar rainfall estimation for flash flood forecasting in small urban watersheds

Radar rainfall estimation for flash flood forecasting in small, urban catchments is examined through analyses of radar, rain gage and discharge observations from the 14.3 km² Dead Run drainage basin in Baltimore County, Maryland. The flash flood forecasting problem pushes the envelope of rainfall estimation to time and space scales that are commensurate with the scales at which the fundamental governing laws of land surface processes are derived. Analyses of radar rainfall estimates are based on volume scan WSR-88D reflectivity observations for 36 storms during the period 2003–2005. Gage-radar analyses show large spatial variability of storm total rainfall over the 14.3 km² basin for flash flood producing storms. The ability to capture the detailed spatial variation of rainfall for flash flood producing storms by WSR-88D rainfall estimates varies markedly from event to event. As spatial scale decreases from the 14.3 km² scale of the Dead Run watershed to 1 km² (and the characteristic time scale of flash flood producing rainfall decreases from 1 h to 15 min) the predictability of flash flood response from WSR-88D rainfall estimates decreases sharply. Storm to storm variability of multiplicative bias in storm total rainfall estimates is a dominant element of the error structure of radar rainfall estimates, and it varies systematically over the warm season and with flood magnitude. Analyses of the 7 July 2004 and 28 June 2005 storms illustrate microphysical and dynamical controls on radar estimation error for extreme flash flood producing storms.     

High spatial resolution mapping of surface velocities and depths for shallow overland flow

Point measurements of flow rate, depth or velocity are not sufficient to validate overland flow models, particularly when the interaction of the water with the soil surface creates a complex flow geometry. In this study, we present the coupling of two techniques obtaining spatial data of flow depths and surface velocity measurements for water depths as low as 1 mm. Overland flow experiments were performed in the laboratory at various flow rates and slopes on two surfaces. The first surface was 120 cm by 120 cm showing three undulations of sinusoidal shape with an amplitude of 1 cm and a wavelength of 20 cm, while the second was a 60 cm by 60 cm moulded reproduction of a seedbed with aggregates up to 2 cm in size. Large scale particle image velocimetry (LSPIV) was used for velocity measurements with a sub‐centimetre spatial resolution. An instantaneous‐profile laser scanner was used to map flow depths with a sub‐millimetre spatial resolution. A sensitivity analysis of the image processing of the LSPIV showed good robustness of the method. Comparison with measurements performed with hot film anemometer and salt velocity gauge showed that LSPIV surface velocities were representative of the flow. Water depths measured with the laser scanner were also in good agreement with single‐point measurements performed with a dial indicator. Spatially‐distributed flow rates could be computed by combining both presented techniques with a mean relative error less than 20%. Copyright © 2012 John Wiley & Sons, Ltd.     

A fast method of flood discharge estimation

Discharge, especially during flood periods, is among the most important information necessary for flood control, water resources planning and management. Owing to the high flood velocities, flood discharge usually cannot be measured efficiently by conventional methods, which explains why records of flood discharge are scarce or do not exist for the watersheds in Taiwan. A fast method of flood discharge estimation is presented. The greatest advantage of the proposed method is its application to estimate flood discharge that cannot be measured by conventional methods. It has as its basis the regularity of open‐channel flows, i.e. that nature maintains a constant ratio of mean to maximum velocities at a given channel section by adjusting the velocity distribution and the channel geometry. The maximum velocity at a given section can be determined easily over a single vertical profile, which tends to remain invariant with time and discharge, and can be converted to the mean velocity of the entire cross‐section by multying by the constant ratio. Therefore the mean velocity is a common multiple of maximum velocity and the mean/maximum velocity ratio. The channel cross‐sectional area can be determined from the gauge height, the water depth at the y‐axis or the product of the channel width multiplied by the water depth at the y‐axis. Then the most commonly used method, i.e. the velocity–area method, which determines discharge as the product of the cross‐sectional area multiplied by mean velocity, is applied to estimate the flood discharge. Only a few velocity measurements on the y‐axis are necessary to estimate flood discharge. Moreover the location of the y‐axis will not vary with time and water stage. Once the relationship of mean and maximum velocities is established, the flood estimation can be determined efficiently. This method avoids exposure to hazardous environments and sharply reduces the measurement time and cost. The method can be applied in both high and low flows in rivers. Available laboratory flume and stream‐flow data are used to illustrate accuracy and reliability, and results show that this method can quickly and accurately estimate flood discharges. Copyright © 2004 John Wiley & Sons, Ltd.     

From manned to unmanned aircraft: Adapting airborne particle size mapping methodologies to the characteristics of sUAS and SfM

Subaerial particle size data holds a wealth of valuable information for fluvial, coastal, glacial and other sedimentological applications. Recently, we have gained the opportunity to map and quantify surface particle sizes at the mesoscale using data derived from small unmanned aerial system (sUAS) imagery processed using structure from motion (SfM) photogrammetry. Typically, these sUAS‐SfM approaches have been based on calibrating orthoimage texture or point cloud roughness with particle size. Variable levels of success are reported and a single, robust method capable of producing consistently accurate and precise results in a range of settings has remained elusive. In this paper, we develop an original method for mapping surface particle size with the specific constraints of sUAS and SfM in mind. This method uses the texture of single sUAS images, rather than orthoimages, calibrated with particle sizes normalised by individual image scale. We compare results against existing orthoimage texture and roughness approaches, and provide a quantitative investigation into the implications of the use of sUAS camera gimbals. Our results indicate that our novel single image method delivers an optimised particle size mapping performance for our study site, outperforming both other methods and delivering residual mean errors of 0.02 mm (accuracy), standard deviation of residual errors of 6.90 mm (precision) and maximum residual errors of 16.50 mm. Accuracy values are more than two orders of magnitude worse when imagery is collected by a similar drone which is not equipped with a camera gimbal, demonstrating the importance of mechanical image stabilisation for particle size mapping using measures of image texture. Copyright © 2017 John Wiley & Sons, Ltd.     

Development of an inexpensive automated streamflow monitoring system

Estimating streamflow is time and labour intensive due to the necessity of developing a rating curve. The development of a rating curve involves acquiring in-field measurements of streamflow across a wide range of flow levels, which can be costly and impractical in remote regions with limited seasonal access. Here we showcase an automated system which estimates streamflow multiple times each day, greatly facilitating the development of rating curves for remote or seasonally inaccessible sites. The system uses an emerging technique referred to as particle image velocimetry (PIV) to track the movement of objects and flow structure features on the mobile water surface to generate velocity vector grids. Velocity grids were used to calculate streamflow and facilitate the development of a rating curve. This work contributes to our field-based knowledge for using automated PIV systems to estimate streamflow in small streams (<5 m wide) using automatically distributed particles.     

Development of new sensor systems for continuous bedload monitoring using a submerged load‐cell system (SLS)

It is important to evaluate bedload discharge and temporal changes of the bed surface, and bed deformation can be estimated during floods if the bedload discharge is properly evaluated in an arbitrary cross‐section. With the exception of grain size and its distribution within the bedload, bedload discharge has been measured using both direct and indirect methods. Bedload slot is a direct method but cannot be used to measure bedload during a flood because of volume limitations. Indirect methods require correlation between the signals and sediment volume measured using another method. In the present study, a small, automatically recording bedload sensor with an iron plate and a pair of load cells is developed in order to evaluate not only large particles but also sand particles as bedload. Bedload mass is calculated by integrating with respect to both the velocity of sediment particles and the averaged particle weight as measured by a pair of load cells, and, as an example, the velocity is estimated by the cross‐correlation function of weights measured by load cells. The applicability of the proposed sensor is discussed based on the results of flume tests in the laboratory (2014) and the observation flume of the Hodaka Sedimentation Observatory of Kyoto University in Japan (2015). The system was installed in the observation flume in November of 2012, and flume data were obtained using natural sediment particles. In particular, it was difficult to estimate the velocity of averaged bedload particles, and it was better to apply a cross‐correlation function in the laboratory tests. However, it appears that the previous estimation can estimate these velocities in the observation flume using a connecting tube and submerged load‐cell systems. Copyright © 2017 John Wiley & Sons, Ltd.     

Biogeochemistry of dissolved carbon,major, and trace elements during spring flood periods on the Ob River

Detailed knowledge of the flood period of Arctic rivers remains one of the few factors impeding rigorous prediction of the effect of climate change on carbon and related element fluxes from the land to the Arctic Ocean. In order to test the temporal and spatial variability of element concentration in the Ob River (western Siberia) water during flood period and to quantify the contribution of spring flood period to the annual element export, we sampled the main channel year round in 2014–2017 for dissolved C, major, and trace element concentrations. We revealed high stability (approximately ≤10% relative variation) of dissolved C, major, and trace element concentrations in the Ob River during spring flood period over a 1‐km section of the river channel and over 3 days continuous monitoring (3‐hr frequency). We identified two groups of elements with contrasting relationship to discharge: (a) DIC and soluble elements (Cl, SO₄, Li, B, Na, Mg, Ca, P, V, Cr, Mn, As, Rb, Sr, Mo, Ba, W, and U) negatively correlated (p < 0.05) with discharge and exhibited minimal concentrations during spring flood and autumn high flow and (b) DOC and particle‐reactive elements (Al, Fe, Ti, Y, Zr, Nb, Cs, REEs, Hf, Tl, Pb, and Th), some nutrients (K), and metalloids (Ge, Sb, and Te), positively correlated (p < 0.05) with discharge and showed the highest concentrations during spring flood. We attribute the decreased concentration of soluble elements with discharge to dilution by groundwater feeding and increased concentration of DOC and particle‐reactive metals with discharge to leaching from surface soil, plant litter, and suspended particles. Overall, the present study provides first‐order assessment of fluxes of major and trace elements in the middle course of the Ob River, reveals their high temporal and spatial stability, and characterizes the mechanism of river water chemical composition acquisition.     

The chronology and the hydrometeorology of catastrophic floods on Dartmoor,South West England

Extreme floods are the most widespread and often the most fatal type of natural hazard experienced in Europe, particularly in upland and mountainous areas. These ‘flash flood’ type events are particularly dangerous because extreme rainfall totals in a short space of time can lead to very high flow velocities and little or no time for flood warning. Given the danger posed by extreme floods, there are concerns that catastrophic hydrometeorological events could become more frequent in a warming world. However, analysis of longer term flood frequency is often limited by the use of short instrumental flow records (last 30–40 years) that do not adequately cover alternating flood‐rich and flood‐poor periods over the last 2 to 3 centuries. In contrast, this research extends the upland flood series of South West England (Dartmoor) back to ca AD 1800 using lichenometry. Results show that the period 1820 to mid‐1940s was characterized by widespread flooding, with particularly large and frequent events in the mid‐to‐late 19th and early 20th centuries. Since ca 1850 to 1900, there has been a general decline in flood magnitude that was particularly marked after the 1930s/mid‐1940s. Local meteorological records show that: (1) historical flood‐rich periods on Dartmoor were associated with high annual, seasonal and daily rainfall totals in the last quarter of the 19th century and between 1910 and 1946, related to sub‐decadal variability of the North Atlantic Oscillation and receipt of cyclonic and southerly weather types over the southwest peninsula; and (2) the incidence of heavy daily rainfall declined notably after 1946, similar to sedimentary archives of flooding. The peak period of flooding on Dartmoor predates the beginning of gauged flow records, which has practical implications for understanding and managing flood risk on rivers that drain Dartmoor. Copyright © 2013 John Wiley & Sons, Ltd.     

Particle velocity and concentration profiles in bedload experiments on a steep slope

Depth profiles of particle streamwise velocity, concentration and bedload sediment transport rate were measured in a turbulent and supercritical water flow. One‐size 6 mm diameter spherical glass beads were transported at equilibrium in a two‐dimensional 10% steep channel with a mobile bed. Flows were filmed from the side by a high‐speed camera. Particle tracking algorithms made it possible to determine the position, velocity and trajectory of a very large number of particles. Approximately half of the sediment transport rate was composed by rolling grains, and the other half by saltation. This revealed a complex structure, with several concentration and flux peaks due to rolling, and one peak due to saltation. With an increase of the sediment transport rate, the depth structure remained the same at the water/granular interface, with peak value increases but with no shift in elevations. The saltation region expanded towards higher elevations with an increase of the particle velocity commensurate to the water velocity. The proportion of the sediment transport rate in saltation did not vary significantly. The particle streamwise velocity profiles exhibited three segments: an exponential decay in the bed, a linear increase where rolling and saltation co‐existed, and above this, a logarithmic‐like shape due to saltating particles. These results are comparable to profiles measured and modelled in dry granular free surface flows and in more intense bedload such as sheet flows. Copyright © 2014 John Wiley & Sons, Ltd.     

Mitigating systematic error in topographic models derived from UAV and ground‐based image networks

High resolution digital elevation models (DEMs) are increasingly produced from photographs acquired with consumer cameras, both from the ground and from unmanned aerial vehicles (UAVs). However, although such DEMs may achieve centimetric detail, they can also display systematic broad‐scale error that restricts their wider use. Such errors which, in typical UAV data are expressed as a vertical ‘doming’ of the surface, result from a combination of near‐parallel imaging directions and inaccurate correction of radial lens distortion. Using simulations of multi‐image networks with near‐parallel viewing directions, we show that enabling camera self‐calibration as part of the bundle adjustment process inherently leads to erroneous radial distortion estimates and associated DEM error. This effect is relevant whether a traditional photogrammetric or newer structure‐from‐motion (SfM) approach is used, but errors are expected to be more pronounced in SfM‐based DEMs, for which use of control and check point measurements are typically more limited. Systematic DEM error can be significantly reduced by the additional capture and inclusion of oblique images in the image network; we provide practical flight plan solutions for fixed wing or rotor‐based UAVs that, in the absence of control points, can reduce DEM error by up to two orders of magnitude. The magnitude of doming error shows a linear relationship with radial distortion and we show how characterization of this relationship allows an improved distortion estimate and, hence, existing datasets to be optimally reprocessed. Although focussed on UAV surveying, our results are also relevant to ground‐based image capture. © 2014 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.