Ice-jam Flooding Research Articles

Ice-jam flood delineation: Challenges and research needs

Identifying and understanding flood hazard and risk is the requisite first step in enhancing public safety, emergency response planning and future flood damage reduction. The delineation and mapping of flood hazard, inundation and risk areas is an effective way to communicate these vulnerabilities. Procedures and standards for flood mapping for open-water (ice-free) events have been performed by many jurisdictions and are well established. In Canada, rivers and streams are prone to ice-related flooding, and these flood events often result in higher water levels and more extensive damages than open-water events do. Ice-related flood events are highly variable and difficult to evaluate, especially with respect to likelihood of site-specific re-occurrence and statistical frequency. In addition, field data and other information required for hydrotechnical modelling of ice-related flood hazard and risk are often unavailable, and standards for ice-related floodplain delineation are less developed. Therefore, many jurisdictions have not adopted guidelines and standards for producing flood hazard and risk mapping incorporating ice-related flood events. In this paper, the present status of delineating ice-related flood events and associated challenges are discussed.

Using Remote Sensing Data to Parameterize Ice Jam Modeling for a Northern Inland Delta

The Slave River is a northern river in Canada, with ice being an important component of its flow regime for at least half of the year. During the spring breakup period, ice jams and ice-jam flooding can occur in the Slave River Delta, which is of benefit for the replenishment of moisture and sediment required to maintain the ecological integrity of the delta. To better understand the ice jam processes that lead to flooding, as well as the replenishment of the delta, the one-dimensional hydraulic river ice model RIVICE was implemented to simulate and explore ice jam formation in the Slave River Delta. Incoming ice volume, a crucial input parameter for RIVICE, was determined by the novel approach of using MODIS space-born remote sensing imagery. Space-borne and air-borne remote sensing data were used to parameterize the upstream ice volume available for ice jamming. Gauged data was used to complement modeling calibration and validation. HEC-RAS, another one-dimensional hydrodynamic model, was used to determine ice volumes required for equilibrium jams and the upper limit of ice volume that a jam can sustain, as well as being used as a threshold for the volumes estimated by the dynamic ice jam simulations using RIVICE. Parameter sensitivity analysis shows that morphological and hydraulic properties have great impacts on the ice jam length and water depth in the Slave River Delta.

Dangerous ice phenomena on the lowland rivers of European Russia

In the Russian climate, the security of populations and economic security are often limited by dangerous ice phenomena. Not only ice-jam floods, but also some processes lead to the violation of operating conditions of various facilities (water intakes, roads, bridges) and damage hydraulic structures and shipping. Currently, rivers’ ice regime characteristics change under the influence of both natural (primarily climatic) and anthropogenic factors. Changes have been analyzed in detail on the basis of observations of 300 hydrological stations in the period from 1936 to 2013. Changes of ice phenomena hazards have been estimated. Hazard assessment of flooding caused by ice jams has been carried out for the Northern Dvina River reach from the Velikiy Ustyug City to Kotlas, the most problematic ice jam flooding area in European Russia. The modeling was performed on the basis of STREAM_2D software complex for current conditions and taking into account the possible construction of various protective structures. Methods for the prevention of negative impact of water at the site are examined.

Interannual kinetics (2010–2013) of large wood in a river corridor exposed to a 50-year flood event and fluvial ice dynamics

Semi-alluvial rivers of the Gaspé Peninsula, Québec, are prone to produce and transport vast quantities of large wood (LW). The high rate of lateral erosion owing to high energy flows and noncohesive banks is the main process leading to the recruitment of large wood, which in turn initiates complex patterns of wood accumulation and reentrainment within the active channel. The delta of the Saint-Jean River (SJR) has accumulated large annual wood fluxes since 1960 that culminated in a wood raft of >3-km in length in 2014. To document the kinetics of large wood on the main channel of SJR, four annual surveys were carried out from 2010 to 2013 to locate and describe >1000 large wood jams (LWJ) and 2000 large wood individuals (LWI) along a 60-km river section. Airborne and ground photo/video images were used to estimate the wood volume introduced by lateral erosion and to identify local geomorphic conditions that control wood mobility and deposits. Video camera analysis allowed the examination of transport rates from three hydrometeorological events for specific river sections. Results indicate that the volume of LW recruited between 2010 and 2013 represents 57% of the total LW production over the 2004–2013 period. Volumes of wood deposited along the 60-km section were four times higher in 2013 than in 2010. Increases in wood amount occurred mainly in upper alluvial sections of the river, whereas decreases were observed in the semi-alluvial middle sections. Observations suggest that the 50-year flood event of 2010 produced large amounts of LW that were only partly exported out of the basin so that a significant amount was still available for subsequent floods. Large wood storage continued after this flood until a similar flood or an ice-breakup event could remobilise these LW accumulations into the river corridor. Ice-jam floods transport large amounts of wood during events with fairly low flow but do not contribute significantly to recruitment rates (ca. 10 to 30% early). It is fairly probable that the wood export peak observed in 2012 at the river mouth, where no flood occurred and which is similar to the 1-in 10-year flood of 2010, is mainly linked to such ice-break events that occurred in March 2012.

Effect of permafrost thaw on the dynamics of lakes recharged by ice-jam floods: case study of Yukon Flats, Alaska

Effect of permafrost thaw on the dynamics of lakes recharged by ice-jam floods: case study of Yukon Flats, Alaska

Impacts of bridge piers on the initiation of ice cover – an experimental study

Abstract Ice jams in northern rivers during winter period significantly change the flow conditions due to the extra boundary of the flow. Moreover, with the presence of bridge piers in the channel, the flow conditions can be further complicated. Ice cover often starts from the front of bridge piers, extending to the upstream. With the accumulation of ice cover, ice jam may happen during early spring, which results in the notorious ice jam flooding. In the present study, the concentration of flowing ice around bridge piers has been evaluated based on experiments carried out in laboratory. The critical condition for the initiation of ice cover around bridge piers has been investigated. An equation for the critical floe concentration was developed. The equation has been validated by experimental data from previous studies. The proposed model can be used for the prediction of formation of ice cover in front of a bridge pier under certain conditions.

Hydroclimatic aspects of ice jam flooding near Perth-Andover, New Brunswick

Major ice jam floods occurred along the Saint John River at Perth-Andover, New Brunswick, in 1976, 1987, 1993, 2009, and 2012. These floods have been devastating and resulted in relocation or demolition of many buildings in affected areas of the community. The purpose of this paper is to review ice processes along the Saint John River from Grand Falls to Beechwood Dam and their relevance to the ice-related problems that have occurred in Perth-Andover. The HEC-RAS program was chosen for ice jam flood stage computations, and cross-sectional, roughness and flow information entered to create a model of the study reach. The model was calibrated using field data obtained on 31 March 2000 along an ice jam in the Perth-Andover area. Model applications resulted in stage-discharge relationships for ice jams of different lengths and locations within the study stretch (Beechwood Dam to Grand Falls). These relationships indicate that ice jams can cause flooding at Perth-Andover, even with moderate spring breakup flows. The variation of ice jamming potential along the study stretch, and its relation to reservoir bathymetry and slope, are discussed. The increase in the number and severity of ice jam floods that have affected Perth-Andover may have resulted from changes in infrastructure and climate that have occurred since the 1950s. Local ice problems are likely to be aggravated over the next few decades but tempered in the longer term if the climate continues to change according to current projections.

Numerical model study on ice impact on Lake Superior outflow limit

Improved regulation of the wintertime flow from Lake Superior is needed to improve the balance of water levels of Lake Superior and Lake Michigan–Huron to decrease the frequency of extreme levels without unduly affecting Lake Superior interest. The wintertime outflow limit is set as 2410 m3/s by Lake Superior Regulation Plan 1977-A as a result of ice jam flooding during the 1916–1917 winter. This paper presents a numerical model study on the ice conditions in the St. Marys River to assess the maximum allowable Lake Superior wintertime outflow. Freeze-up, frazil transport and accumulation, and breakup were simulated with a thermal-ice dynamic model. The highest potential for flooding exists during ice breakup and simulations were run to determine a safe discharge limit for the breakup period. Simulations indicated the winter flow limit may be increased to 2690 m3/s if flow regulation is managed with care to prevent premature ice cover breakup.

Long lead forecasting of spring peak runoff using Mamdani-type fuzzy logic systems at Hay River, NWT

In northern riverside communities, breakup ice jam flooding is an annual threat to properties and human lives. In this study, the peak snowmelt runoff during breakup was assessed as an indicator of breakup flood severity. Due to the sparse network of hydrometeorolocial data in remote northern regions, a Mamdani-type fuzzy logic system (FLS) was developed and tested with the limited historical data. Three input variables were defined from the precipitation, air temperature and daily water level data. All of these variables are known ∼3 to 4 weeks before breakup enabling a long lead-time forecast. The process of system development is demonstrated by a case study of the Town of Hay River, NWT Canada. A series of experiments were designed to select the best system configuration, which also provided a way to conduct a sensitivity analysis for different choices in each system component. It was found that the system shows very good performance on the historical data using the qualitative error index. The results of the sensitivity analysis suggest the system performance is dependent on the choices of fuzzy logic inference operators and defuzzification method. As a long lead system, the short-term meteorological factors that would affect the system output were analyzed and the possible error range was assessed. Preliminary model validation, based on three years of testing, shows promising performance.

Interdisciplinary approach to hydrological hazard mitigation and disaster response and effects of climate change on the occurrence of flood severity in central Alaska

Abstract. In May 2013, a massive ice jam on the Yukon River caused flooding that destroyed much of the infrastructure in the Interior Alaska village of Galena and forced the long-term evacuation of nearly 70% of its residents. This case study compares the communication efforts of the out-of-state emergency response agents with those of the Alaska River Watch program, a state-operated flood preparedness and community outreach initiative. For over 50 years, the River Watch program has been fostering long-lasting, open, and reciprocal communication with flood prone communities, as well as local emergency management and tribal officials. By taking into account cultural, ethnic, and socioeconomic features of rural Alaskan communities, the River Watch program was able to establish and maintain a sense of partnership and reliable communication patterns with communities at risk. As a result, officials and residents in these communities are open to information and guidance from the River Watch during the time of a flood, and thus are poised to take prompt actions. By informing communities of existing ice conditions and flood threats on a regular basis, the River Watch provides effective mitigation efforts in terms of ice jam flood effects reduction. Although other ice jam mitigation attempts had been made throughout US and Alaskan history, the majority proved to be futile and/or cost-ineffective. Galena, along with other rural riverine Alaskan communities, has to rely primarily on disaster response and recovery strategies to withstand the shock of disasters. Significant government funds are spent on these challenging efforts and these expenses might be reduced through an improved understanding of both the physical and climatological principals behind river ice breakup and risk mitigation. This study finds that long term dialogue is critical for effective disaster response and recovery during extreme hydrological events connected to changing climate, timing of river ice breakup, and flood occurrence in rural communities of the Far North.

Imitation modeling of ice dams (case study of Tom’ River, Western Siberia)

The factors of ice jam formations in the lower flow of the Tom River (Siberia) are investigated. A length of the main channel under investigation is about 120 km. Approaches to solution of the problem of the jam formation control and, as a consequence, the jam induced floods are considered on the basis of the imitative computer modeling of stream dynamics and ice jams. The simulation makes it possible to analyze different scenarios of initial forcing and to predict reactions of the river bed system to the effects. On the basis of 1D models developed in the HEC-RAS 4.0 modeling system for the Tom River at the city of Tomsk we investigated a possibility of the ice jam localization, probability of which at different parts of river flow varies in time according to change of the river water discharge, stream hydraulics, and ice cover thickness. The 2D hydrodynamic model of the Tom River channel system in the SMS 9.2 modeling system has been developed. It allows simulating effects of ice jams located in different sections of the river flow on the run-off redistribution between the main channel and other river branches. It makes possible to estimate hazards and risks of ice jam floods and probable effects of ice jams on formation of the river channel system. As a result it becomes possible to regulate the safe spring ice transit through populated areas. Analysis of factors of the ice jam formations has demonstrated that due to increasing anthropogenic influence changes of hydro-meteorological and geomorphologic conditions lead to more frequent occurrence of jam floods for the last 25 years as compared to previous 40-year period. The imitative computer models are proposed to be used for planning anti-jam measures since they make possible to create a whole system of the channel structure, a relief of channel and floodplain, a flow velocity field including dangerous hydrologic processes. Similar system would allow predicting both consequences of local anthropogenic influences and their synergic effects. A change of a pressure head along the river course under fixed water discharges at initial hydraulic section can be calculated with the 1D flow model so that to determine places of the most probable localization of the ice jams. A computer simulation of jams in the river channel system with use of preassigned increased roughness coefficients allows quantitative estimating of the runoff redistribution within the channel system under certain scenarios of a spring flood in the jam-dangerous parts of branching river channels. It is important to preserve such branches, and they should be cut by any dam from a main channel.

Simulations of ice jam thickness distribution in the transverse direction

River ice often forms in the cold regions of northern hemisphere which can lead to ice jams (or ice dams). Water level can be significantly raised due to ice jams. As a consequence, disastrous ice flooding may be resulted, such as the ice jam flooding in the Nechako River in Prince George in winter 2007-2008. In the present study, the equations describing the ice jam thickness in the transverse direction are derived. The impact of the secondary vortex is considered while the cohesive force within ice cubes is neglected in the model. The relationship between the parameter β and the total water depth is established based on the assumption that all other variables except the velocities are kept constant on the same cross section. By using the parameter β and the developed equations, the ice jam thickness in the transverse direction can be predicted. The developed model is used to simulate the ice jam thickness in the transverse direction at the Hequ Reach of the Yellow River in China. The simulated ice jam thicknesses agree well with the field measurements on different cross sections.

Comparing the impacts of regulation and climate on ice-jam flooding of the Peace-Athabasca Delta

The Peace-Athabasca Delta (PAD) in northern Alberta is one of the world's largest inland freshwater deltas, home to large populations of waterfowl, muskrat, beaver, and free-ranging wood bison. Following construction of the Bennett Dam in the late 1960s, a paucity of ice-jam flooding in the lower Peace River has resulted in prolonged dry periods and considerable reduction in the area covered by lakes and ponds that provide habitat for aquatic life in the PAD region. Past studies have shown that both regulation and climate have contributed to the drying trend, via increased freezeup levels and reduced breakup flows, respectively. However, it has not so far been possible to assess the relative impacts of these two factors. This question is addressed herein by developing quantitative relationships based on the concept of conditional probability. The new methodology is first tested against observed frequencies of ice-jam floods both before and after regulation; it is subsequently applied to quantify the effects of climate and of regulation on the post-regulation reduction in ice-jam flood frequency. The results indicate that both factors have contributed significantly to the drying of the PAD, with regulation having had the more pronounced effect. Complications that may arise from secondary effects of regulation on breakup flows or of climate on freezeup levels are discussed and quantified. It is shown that such effects do not alter the overall conclusions of the paper.

Ontogeny of point bars on a river in a cold semi-arid climate

During the extreme flood of May 1978 in Powder River, Montana, USA, three new point bars were created: two by meander cutoffs, and one by 65 m of lateral channel migration. By using annual cross-sectional surveys and by sampling sediment in dug trenches, we documented the creation of point-bar platforms upon which sediments were later deposited to create point-bar features that evolved with time. Our observations of point-bar growth and evolution on Powder River are compared with similar observations of other point bars in humid subtropical and subarctic climates. Powder River is a meandering alluvial river in a cold semi-arid climate, characterized by a diverse flow regime that includes occasional ice-jam floods in late winter, annual snowmelt floods in late spring, episodic flash floods in the summer, and infrequent floods in the fall. The building of the point-bar platform took place during a relatively short time span, by an apparently random process of deposition and erosion of unit bars during floods that gradually changed a concave channel surface to a convex surface upon which the point-bar features evolved. Point-bar features on Powder River were built by a superposition of multiple unit bars associated with five flood types over a period of years rather than during a single year. Erosion was found to be a significant process in the shaping of the point bars. Erosion varied spatially between point bars and temporally at multiple time scales over the 33 yr time span. Erosion at the decadal time scale indicated that on average 19%–41% of the initial deposits (older than 10 yr) were eroded. This quantification of erosion has implications for interpreting the geological record.

Temporal and spatial pattern of thermokarst lake area changes at Yukon Flats, Alaska

To better understand the linkage between lake area change, permafrost conditions and intra-annual and inter-annual variability in climate, we explored the temporal and spatial patterns of lake area changes for a 422 382-ha study area within Yukon Flats, Alaska using Landsat images of 17 dates between 1984 and 2009. Only closed basin lakes were used in this study. Among the 3529 lakes greater than 1 ha, closed basin lakes accounted for 65% by number and 50% by area. A multiple linear regression model was built to quantify the temporal change in total lake area with consideration of its intra-annual and inter-annual variability. The results showed that 80.7% of lake area variability was attributed to intra-annual and inter-annual variability in local water balance and mean temperature since snowmelt (interpreted as a proxy for seasonal thaw depth). Another 14.3% was associated with long-term change. Among 2280 lakes, 350 lakes shrank, and 103 lakes expanded. The lakes with similar change trends formed distinct clusters, so did the lakes with similar short term intra-annual and inter-annual variability. By analysing potential factors driving lake area changes including evaporation, precipitation, indicators for regional permafrost change, and flooding, we found that ice-jam flooding events were the most likely explanation for the observed temporal pattern. In addition to changes in the frequency of ice jam flooding events, the observed changes of individual lakes may be influenced by local variability in permafrost distributions and/or degradation. Copyright © 2012 John Wiley & Sons, Ltd.

Historic floods in the city of Prague – a reconstruction of peak discharges for 1481–1825 based on documentary sources

The paper aims to reconstruct peak discharges of historic floods in an urbanized area of the historic city of Prague based on documentary sources from pre-instrumental and the early instrumental period (1481–1825). Approximately 20–30 maximum water levels are denoted by flood-marks, accounts describing or related at unchanged sites, or by early instrumental measurements. The challenge in this reconstruction is the identification and consideration of man-made floodplain modifications influencing the cross-section area and the hydraulic roughness. In order to overcome this problem, a simple approach to estimate peak discharges of historic floods has been developed and applied to the River Vltava. This approach includes a procedure for reconstructing the hydraulic parameters of the river channel and inundated floodplain, coupled with an approach for the verification of estimated peak discharge reliability. As a result of the different hydraulic characteristics associated with ice jam floods all winter-flood events are excluded to avoid their potential inclusion. We present 18 reconstructed discharge maxima. The validation of the technique by comparison with the recent gauged flood of 2002 reveals results of adequate accuracy. The comparison also shows that the flood event of 2002 was conspicuously greater than all calculated summer floods in 1481–1825.

Developing temporal hydroecological perspectives to inform stewardship of a northern floodplain landscape subject to multiple stressors: paleolimnological investigations of the Peace–Athabasca Delta

Effective stewardship of ecologically-significant floodplain landscapes requires knowledge of the relative roles of natural processes and upstream human activities on environmental flows. In these landscapes, hydroecological conditions that develop from potentially competing drivers, such as climate change and industrial development, tend to be expressed at spatial and temporal scales that are often inadequately captured by existing monitoring datasets. Consequently, perceived cause–effect relations may be misunderstood, conflict can escalate among stakeholders, and effectiveness of surveillance systems, policies, and governance may be impaired. This is the context for the Peace–Athabasca Delta (PAD), an internationally-recognized water-rich floodplain landscape located in northern Alberta (Canada) that has been subject to multiple stressors. Here we synthesize evidence from paleolimnological records that have fostered an unparalled window into the natural history of this landscape. Over the past 12 years, we have assembled numerous decadal- to multicentennial-long records of hydrological and ecological variability, including an exceptionally detailed chronicle of Peace River flood frequency and magnitude spanning ~600 years. These efforts recently culminated in a 5200-year reconstruction of Lake Athabasca water-level history. Results have provided the foundation to identify drivers of landscape change and generate insight into the delta’s dynamic and ongoing evolution. Contrary to widespread perceptions that hydroelectric regulation of the Peace River since the late 1960s has reduced the frequency of ice-jam floods and lowered floodplain lake-water levels, results indicate that climate variability exerts the overwhelming influence on the delivery of water to the PAD. We show that impending climate-driven freshwater scarcity of a scale unprecedented in our collective societal memory now poses a significant threat to the ecological integrity of this world-renowned landscape and a major challenge to water resource managers. Also, we propose a hydroecological monitoring program, built upon the knowledge gained from our extensive process studies and paleoenvironmental research, to inform effective ongoing stewardship of the delta.

Hydrodynamic characteristics and effects of river waves caused by ice jam releases

Abstract Ice-jam release waves, or javes for short, are highly dynamic events that occur in ice-forming rivers. They are known for serious ecological and socio-economic impacts, which are usually detrimental but on occasion beneficial, particularly with respect to ecosystem maintenance in the large freshwater deltas of northern Canada. Detailed water level data have been obtained during the passage of javes in the past 10 years or so, but their study is hampered by lack of velocity and discharge measurements, owing to the presence of moving ice in the flow. A previously developed method to analytically compute such parameters, which is based on measured waveforms, has given encouraging results in several applications. Herein, this method is first subjected to comprehensive validation using numerical model output for a hypothetical jave in a prismatic channel. The analytical method is shown to perform adequately with respect to all hydrodynamic jave characteristics, such as celerity, velocity, discharge, and bed shear stress. Jave data from various rivers are then compiled in tabular form and it is shown via dimensional analysis that bulk predictions can be made in terms of jave height and rate of rise of the water level. Implications of the present findings to riverbed scour and to ecologically vital ice-jam flooding of flat rivers and deltas are explored and research challenges identified.

Applicability of 2-D modeling for forecasting ice jam flood levels in the Hay River Delta, Canada

Ice jam floods can present an annual threat to communities adjacent to rivers, especially those situated at river confluences, island or deltas. The objective of this study was to determine whether 2-D modeling might be used to predict expected ice jam flood elevations for such multi-channel systems. The Hay River delta was selected as the demonstration site for this study, and 2-D modeling was employed to calculate ice jam flood levels expected for varying inflow discharge conditions. The River2D model was calibrated for historical ice jam profiles covering a range of discharges from 268 to 1000 m3/s and a relationship was developed to predict the flow split down the East and West Channels based on the inflow discharge to the delta. Also, ice jam rating curves were developed at 0.5 km intervals along each channel facilitating the development of an ice jam profile prediction tool for use by the Town of Hay River.

Measurements and analysis of ice breakup and jamming characteristics in the Mackenzie Delta, Canada

Ice breakup is a controlling factor in the hydrology of arctic deltas, including the Mackenzie River Delta, which is characterized by a flat front and numerous channels and lakes. Ice-jam flooding replenishes delta lakes with essential water, sediment and nutrients. The present study, carried out under the auspices of the International Polar Year, aims to gather essential quantitative information for improving understanding of delta ice processes and associated prediction capability. Comprehensive observations and measurements, which were carried out during the 2007 and 2008 breakups, are described along with relevant analysis and data interpretation. The spacing of pre-breakup transverse cracks in delta channels is similar to that of rivers, typically ranging from 2 to 5 ice cover widths. Ice jamming in the delta is driven by major ice runs originating in the Mackenzie River and stalling in the upper reaches of Middle Channel. The data collected during the mechanical breakup event of 2008 permitted crude, but first-ever, quantification of flow re-distribution resulting from the formation of a Middle Channel jam. This jam also generated jams in East Channel and smaller distributaries. Examination of threshold velocity for ice block submergence indicated that, with notable exceptions, the expected kind of jam in delta channels is a surface accumulation of ice blocks and plates. This prediction is in accord with observations and hydrometric data at the Inuvik and Aklavik gauges. The use of portable loggers to record water level variations at selected sites proved to be a robust and effective technique for obtaining key hydraulic data.