Hydrometeorological Characteristics of Ice Jams on the Pemigewasset River in Central New Hampshire

Matthew C. Sanders Meteorology Program, Plymouth State University, Plymouth, New Hampshire

Search for other papers by Matthew C. Sanders in
Current site
Google Scholar
PubMed
Close
,
Jason M. Cordeira Meteorology Program, Plymouth State University, Plymouth, New Hampshire

Search for other papers by Jason M. Cordeira in
Current site
Google Scholar
PubMed
Close
, and
Nicholas D. Metz Department of Geoscience, Hobart and William Smith Colleges, Geneva, New York

Search for other papers by Nicholas D. Metz in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Ice jams that occurred on the Pemigewasset River in central New Hampshire resulted in significant localized flooding on 26 February 2017 and 13 January 2018. Analyses of these two case studies shows that both ice jam events occurred in association with enhanced moisture transport characteristic of atmospheric rivers (ARs) that resulted in rain-on-snow, snowpack ablation, and rapid increases in streamflow across central New Hampshire. However, while the ice jams and ARs that preceded them were similar, the antecedent hydrometeorological characteristics of the region were different. The February 2017 event featured a “long melting period with low precipitation” scenario, with several days of warm (~5°–20°C) maximum surface temperatures that resulted in extensive snowmelt followed by short-duration, weak AR that produced ~10–15 mm of precipitation during a 6-h period prior to the formation of the ice jam. Alternatively, the January 2018 event featured a “short melting period with high precipitation” scenario with snowmelt that occurred primarily during a more intense and long-duration AR that produced >50 mm of rainfall during a 30-h period prior to the formation of the ice jam. Composite analysis of 20 ice jam events during 1981–2019 illustrates that 19 of 20 events were preceded by environments characterized by ARs along the U.S. East Coast and occur in association with a composite corridor of enhanced integrated water vapor > 25 mm collocated with integrated water vapor transport magnitudes > 600 kg m−1 s−1. Additional analyses suggest that most ice jams on the Pemigewasset River share many common synoptic-scale antecedent meteorological characteristics that may provide situational awareness for future events.

© 2020 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Jason M. Cordeira, j_cordeira@plymouth.edu

Abstract

Ice jams that occurred on the Pemigewasset River in central New Hampshire resulted in significant localized flooding on 26 February 2017 and 13 January 2018. Analyses of these two case studies shows that both ice jam events occurred in association with enhanced moisture transport characteristic of atmospheric rivers (ARs) that resulted in rain-on-snow, snowpack ablation, and rapid increases in streamflow across central New Hampshire. However, while the ice jams and ARs that preceded them were similar, the antecedent hydrometeorological characteristics of the region were different. The February 2017 event featured a “long melting period with low precipitation” scenario, with several days of warm (~5°–20°C) maximum surface temperatures that resulted in extensive snowmelt followed by short-duration, weak AR that produced ~10–15 mm of precipitation during a 6-h period prior to the formation of the ice jam. Alternatively, the January 2018 event featured a “short melting period with high precipitation” scenario with snowmelt that occurred primarily during a more intense and long-duration AR that produced >50 mm of rainfall during a 30-h period prior to the formation of the ice jam. Composite analysis of 20 ice jam events during 1981–2019 illustrates that 19 of 20 events were preceded by environments characterized by ARs along the U.S. East Coast and occur in association with a composite corridor of enhanced integrated water vapor > 25 mm collocated with integrated water vapor transport magnitudes > 600 kg m−1 s−1. Additional analyses suggest that most ice jams on the Pemigewasset River share many common synoptic-scale antecedent meteorological characteristics that may provide situational awareness for future events.

© 2020 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Jason M. Cordeira, j_cordeira@plymouth.edu

1. Introduction

Ice jams are common cold-season events in many mid- and high-latitude countries around the world (Rokaya et al. 2018) that affect up to 60% of river reaches in the Northern Hemisphere (Bennett and Prowse 2010). Multiple ice jam case studies reveal that North America is one of the most ice jam prone regions in the world (Rokaya et al. 2018) and these ice jams can form on many rivers in the U.S. Northeast (Fig. 1a). Ice jams form when ice floes accumulate across a river channel and block the normal flow of water downstream. The jam can form at natural bends, constrictions, and areas of slower streamflow in the river channel, and at structures such as bridges and docks. Ice jams can result in flooding upstream of where the jam forms (i.e., the jam front), especially during periods of enhanced streamflow, or downstream of the jam front if the jam breaks. The total cost of ice jams and associated flooding in North America is estimated at $300 million (USD) per year (French 2018) and often burdens rural smaller towns, such as the January 1984 event in Salmon, Idaho, that resulted in $1.8 million (1984 USD) in damages (Zufelt and Bilello 1992) and the January 2018 events across the U.S. Northeast that resulted in flood-related evacuations of 280 people in Pennsylvania, Vermont, and New York (FloodList 2018).

Fig. 1.
Fig. 1.

(a) Locations of all recorded ice jams during 1780–2019 (white circles) and during 1981–2019 (blue circles) across New York and New England in the CRREL Ice Jam Database with the location of the study region indicated with the black box. (b) Aerial photograph of the February 2017 ice jam event looking south illustrating flooding in Holderness, NH. (c) Overview map of the study region indicating the locations of surface meteorological sites (white “+”), USGS river gauges (black diamonds), and the location of the CFS grid point (white “×”) used in the analyses. The red star marks the approximate location of the photograph in (b). The maps in (a) and (c) were created at using ESRI ArcMap 10.5.1 and the image in (b) was obtained from https://www.usgs.gov/media/images/flooding-caused-ice-jam-pemigewasset-river.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

This study is motivated by two ice jams that occurred on the Pemigewasset River in central New Hampshire in February 2017 and January 2018 that resulted in localized yet significant flooding in the towns of Plymouth and Holderness (Fig. 1b). For example, the February 2017 ice jam resulted in flood waters that overtopped Route 175A, closed exits on Interstate 93, inundated two gas stations, closed buildings on the Plymouth State University campus, necessitated the evacuation of local housing, and flooded a Plymouth State University parking lot, submerging ~50 student vehicles (WMUR 2017). According to the Town of Plymouth Hazard Mitigation Plan (Town of Plymouth New Hampshire 2016), riverine and ice jam flooding are a “moderate” severity and “probable” (defined as a 50%–99% likelihood over the next 25 years) occurrence that results in the highest natural hazard risk alongside severe winter weather to the region. This study therefore describes the hydrometeorological characteristics of this natural hazard risk via case studies of the 2017 and 2018 ice jam events and via composite analysis of 20 ice jam events on the Pemigewasset River in central New Hampshire for 1981–2019.

Previous research on ice jams primarily focuses on the mechanics associated with different types of ice jams (e.g., breakup, freeze-up, grounded, and floating ice jams; e.g., FEMA 2018) and modeling their formation and break up (e.g., Rokaya et al. 2018, and references therein). Breakup ice jams typically form during the second half of winter as river ice breaks off the river bed, fractures, and accumulates downstream due to enhanced streamflow and increased discharges associated with warming temperatures, snowmelt, and rainfall (e.g., Bates and Brown 1982). Freeze-up ice jams typically form during the first half of winter as frazil ice accumulates in the river (e.g., Zufelt and Bilello 1992), whereas grounded ice jams form when ice attached to the river bed diverts flow over the banks (e.g., Beltaos and Wong 1986) and floating ice jams form on larger rivers where water is forced to flow around and under surface ice. In this study, we primarily focus on breakup ice jams, which occurred during February 2017 and January 2018 on the Pemigewasset River, as they often pose the largest threat to nearby communities given increased discharges and larger volumes of ice relative to other types of ice jams. For example, the U.S. Army Corps of Engineers (USACE) Engineer Research and Development Centers (ERDC) Cold Regions Research and Engineering Laboratory (CRREL) typically gauges the potential for breakup type ice jams by noting that the increased discharge must increase the river stage by 1.5–3 times the ice thickness in order to produce sufficient mechanical force to fracture riverine ice (UCAR 2020).

Limited previous research has examined the synoptic conditions associated with ice jams. Given that there is no rigorous theoretical framework for predicting ice jam events, the selection of variables to include in an ice jam prediction model can be somewhat arbitrary and include hydrometeorological inputs such as daily minimum, daily maximum, and daily average temperatures, accumulated thawing and freezing degree days, ice thickness, precipitation, basin discharge, accumulated snow cover, and solar radiation (Madaeni et al. 2020). The regional frequency of ice jams may be influenced by both seasonal and intraseasonal variability in the large-scale atmospheric flow pattern over North America. These atmospheric flow patterns may influence the prolonged formation of riverine ice due to a transition to winter conditions or sudden formation of riverine ice due to a short-duration cold snap. Alternatively, breakup of riverine ice may occur due to springtime melt or a short-duration midwinter thaw. For example, a primary factor that determines ice breakup in the interior of Alaska during April–May is warming surface temperatures (Bieniek et al. 2011) that often occurs during El Niño years when fewer storms traverse Alaska and more sunshine is present. Over North America, ice breakup dates appear to correlate best with surface temperatures in the month prior to an event (Palecki and Barry 1986; Robertson et al. 1992; Magnuson et al. 2000). Further, breakup ice jams on the Peace River in Alberta, Canada, have occurred with large-scale flow patterns supporting above normal monthly precipitation (Romolo et al. 2006a) and increased monthly snowmelt (Romolo et al. 2006b).

Abrupt variability in the large-scale flow patterns (i.e., regime transitions) that may trigger the onset of precipitation and temperature extremes that can enhance snowmelt and streamflow within a watershed is also important to ice jam formation. For example, a regime transition in January 1996 led to amplified atmospheric flow and the poleward transport of a warm, humid air mass in the warm sector of midlatitude cyclones over eastern North America. This flow pattern and cyclone produced 72-h precipitation totals of 30–90 mm, melted snow water equivalent (SWE) of 7–15 cm, and led to ice jams and flooding in Pennsylvania (Leathers et al. 1998). Similar regime transitions and synoptic-scale flow patterns occurred in conjunction with 24-h precipitation totals > 30 mm and melted SWE > 10 cm that produced ice jams and flooding in central Vermont in January 1976 and March 1977 (Bates and Brown 1982). The large-scale flow patterns during these two events are examples similar to the Maddox et al. (1979) “synoptic” type extreme precipitation event that occurs downstream of eastern U.S. troughs that may occur in association with a combination of southerly moisture transport and strong synoptic-scale quasigeostrophic forcing for ascent (e.g., Agel et al. 2019; Barlow et al. 2019). Similarly, Lapenta et al. (1995) identify that extreme precipitation in the warm sector of extratropical cyclones along with southerly moisture transport from the Atlantic Ocean and/or Gulf of Mexico and orographic ascent is linked to floods in the Northeast.

The synoptic-scale flow patterns described above share similarities with characteristics that are commonly described by atmospheric rivers (ARs). ARs are long and narrow corridors of enhanced atmospheric integrated water vapor (IWV) and IWV transport (IVT) in midlatitude regions that often accompany cyclones within the midlatitude storm track (e.g., Newell et al. 1992; Ralph et al. 2004; Ramos et al. 2015). The IVT along an AR occurs in association with water vapor flux along the pre-cold-frontal low-level jet of midlatitude cyclones (Ralph et al. 2005) that may intersect orographic barriers and result in enhanced precipitation (e.g., Neiman et al. 2002; Smith et al. 2010; Hecht and Cordeira 2017). The ARs may also result in extreme precipitation where water vapor flux persists in regions of complex terrain (e.g., Lamjiri et al. 2017) and are often characterized by elevated freezing levels which leads to a greater frequency of liquid precipitation and the possibility of rain-on-snow events (e.g., Guan et al. 2016; Ralph et al. 2019). These rain-on-snow events may therefore lead to enhanced streamflow and significant flooding events such as what occurred during the Oroville Dam spillway incident in California during February 2017 (White et al. 2019).

ARs occur in association with a large majority of high-impact precipitation events that lead to flooding on the U.S. West Coast (e.g., Ralph et al. 2006; Guan et al. 2010; Dettinger et al. 2011; Dettinger 2013; Rutz et al. 2014; Ralph et al. 2016; Barlow et al. 2019) and also occur in association with high-impact precipitation events and flooding on the U.S. East Coast, notably over the Southeast (e.g., Moore et al. 2012; Mahoney et al. 2016; Miller et al. 2018). The relationship among ARs, precipitation, flooding, and ice jams over the Northeast has not been quantified; however, the ingredients used to describe ARs are well-known ingredients used by meteorologists within the National Weather Service and Northeast U.S. River Forecast Center to forecast the potential for extreme precipitation and flooding across the region (e.g., Lapenta et al. 1995; J. Arnott, J. Dellicarpini, and D. Vallee 2020, personal communications). Irrespective of ARs, winter and vernal flooding over the Northeast is complicated by hydrometeorological processes that include frozen soils, snow cover, and riverine ice that may subsequently influence streamflow (e.g., Graybeal and Leathers 2006; Pradhanang et al. 2013). For example, rainfall events that produce snowmelt and increased streamflow during mid-to-late winter may result in ice jam formation and flooding that might not occur in early winter in the absence of an antecedent snowpack or riverine ice.

The objectives of this study are twofold: 1) illustrate the hydrometeorological characteristics of two ice jam events via case studies from February 2017 and January 2018 and 2) illustrate the hydrometeorological characteristics of 20 ice jam events via composite analysis for 1981–2019, both focusing on the Pemigewasset River in central New Hampshire. This study will use also adopt an AR perspective to describe the synoptic-scale environment that resulted in antecedent precipitation and snowmelt. It is hypothesized that the common ingredients to ARs mentioned above are not likely to always result in an ice jam owing to necessary antecedent conditions (i.e., the presence of river ice), but that observed ice jams may frequently occur in association with the common ingredients to ARs. The rest of the paper is as follows. Section 2 outlines the datasets and methods used in this study. Section 3 discusses the February 2017 and January 2018 case studies in detail, whereas section 4 presents a climatology of 20 ice jam events that occurred on the Pemigewasset River in central New Hampshire. Section 5 provides the conclusions from this study and suggestions for future work.

2. Data and methods

Ice jams are identified from an “Ice Jam Database” maintained by the USACE ERDC CRREL. The Ice Jam Database was created in 1992 to address the “lack of systematically compiled data on ice events” that “hampered effective ice jam emergency response and hindered research and development” (Weyrick et al. 2007). The Ice Jam Database contains >22 000 ice jams in the United States since 1780, including 4660 ice jams in the Northeast (Fig. 1a). More recently, the modern reanalysis period of 1981–2019 contains 1847 ice jams in the Northeast (Fig. 1a), >250 ice jams in New Hampshire, and specifically 18 events on the Pemigewasset River at Plymouth, New Hampshire. Each entry in the Ice Jam Database includes date of event, location, nearby river gauge stations, hydrologic unit codes, damages, and a record contact, among other fields (USACE 2019). The Ice Jam Database is compiled from reports published by the United States Geological Survey (USGS), historical newspaper reports, and various other sources. The Ice Jam Database contains only recorded ice jam events, but other ice jams have occurred for which no records exist in the database or for those yet to be recorded in the database (White and Eames 1999). For example, the ice jams that led to localized flooding on the Pemigewasset River in Plymouth, New Hampshire, in January 2018 and January 2019 are not yet included in the Ice Jam Database as of July 2020. This study therefore focuses on the 20 recent known ice jam events (i.e., the 18 ice jams events in the Ice Jam Database and the two more recent aforementioned ice jam events) on the Pemigewasset River at Plymouth, New Hampshire, that occurred between January 1981 to February 2019 (Table 1). It is likely that other less impactful ice jam events, perhaps that did not lead to localized flooding, have occurred and were not recorded.

Table 1.

Dates and characteristics of 20 ice jams on the Pemigewasset River in Plymouth, New Hampshire, 1981–2019. Characteristics include the time of maximum IVT magnitude tmax and the maximum IVT magnitude (kg m−1 s−1) derived from the CFS grid point at 44.0°N, 71.5°W, daily snow depth decrease (cm) at PHM, valley and mountain temperature T (°C) and dewpoint temperature Td (; °C) at tmax, and mean areal 48-h precipitation (mm). Valley observations were selected from K1P1, KLCI, and KCON based on availability of observations. Mountain observations at KMWN marked with an asterisk are observations from ±3 h of the corresponding tmax due to missing observations at tmax. The mean areal precipitation represents the PRISM-derived 48-h accumulations associated with each event for a box over the western White Mountains shown in Fig. 12a.

Table 1.

To examine the large-scale meteorological conditions associated with ice jams over the Pemigewasset River, this study uses gridded data from the National Oceanic and Atmospheric Administration (NOAA) National Centers for Environmental Prediction (NCEP) Climate Forecast System (CFS). The CFS is available as a reanalysis dataset for the period from 1 January 1979 to 31 March 2011 (Saha et al. 2006) and as an operational analysis dataset from 1 April 2011 to the present day (Saha et al. 2014). In both versions, the CFS gridded data are available four times daily and with 0.5° latitude × 0.5° longitude horizontal grid spacing. In addition to native CFS meteorological parameters (e.g., IWV as total column precipitable water), this study derives IVT following the methodology of Neiman et al. (2008) and Moore et al. (2012) from the NCEP Climate Forecast System Reanalysis (CFSR) as
IVT=1gpbpt(qVh)dp,
where q is the specific humidity, Vh is the horizontal wind vector, pb is 1000 hPa, pt is 300 hPa, and g is the acceleration due to gravity. Common IWV and IVT magnitude thresholds used for identifying ARs from gridded reanalysis and forecast data over the northeast Pacific often include a combination of IWV values ≥ 20 mm and IVT magnitudes ≥ 250 kg m−1 s−1 as discussed in Rutz et al. (2014) and Cordeira et al. (2017). The IVT, however, is more practical in use as compared to IWV to better emphasize the transport of water vapor and its role in upslope water vapor flux and orographic precipitation. For example, the daily average IVT magnitude explains ~50% of the variance in 24-h precipitation across the western United States, whereas the IWV explains ~25% of that variance (Rutz et al. 2014). The IVT is therefore the most common ingredient used to track and detect ARs globally, with regionally varying criteria and thresholds depending on methodology, that emphasize corridors or regions of enhanced or climatologically significant water vapor transport (Shields et al. 2018).

Composite analyses in this study used both the CFS and the NCEP–National Center for Atmospheric Research (NCAR) reanalysis (Kalnay et al. 1996); the latter is available four times daily with 2.5° latitude × 2.5° longitude horizontal grid spacing and is used primarily as a temporally consistent reanalysis in order to assesses departures from climatology. These data were used to analyze departures from climatology of 500-hPa geopotential height and 850-hPa temperature during selected periods prior to and during each ice jam event. Averaging periods are derived from statistically significant departures in the observed daily average surface temperatures at Pinkham Notch (PHM), New Hampshire, as discussed in section 4. Statistical significance of the reanalysis and PHM analyses is calculated using a two-sided Student’s t test at the 90%, 95%, and 99% confidence levels [Wilks 2006, section 5.2.1] as compared to a 30-yr 1979–2008 climatology following the method of Hart and Grumm (2001). Additional metrics and illustrations display the 48-h mean areal precipitation for each event in Table 1 derived from the 4-km gridded Parameter-Elevation Regressions on Independent Slopes Model (PRISM; Daly et al. 1994) for a domain encompassing 43.5°–44.5°N and 72.0°–71.0°W. Herein we use the gridded PRISM quantitative precipitation estimate reanalysis product owing to limited availability of observed precipitation across the rural and mountainous study domain, despite its potentially limited reliability (e.g., Bosilovich et al. 2008).

The study also uses a variety of surface-based hydrometeorological data at different locations across a subregion of New Hampshire (i.e., “the study region”; Fig. 1c) obtained from the NOAA National Centers for Environmental Information (NCEI 2019). These data include hourly temperature, dewpoint temperature, and liquid-equivalent precipitation at the “valley locations” of Plymouth (K1P1; 150 m MSL), Laconia (KLCI; 166 m MSL), and Concord (KCON; 104 m MSL), and at a “mountain location” of the Mount Washington Observatory (KMWN; 1918 m MSL). Valley observations are prioritized at K1P1 where data are available for 2005–present. Valley observations for data prior 2005 or for any missing data during 2005–present are from KLCI; any missing observations at KCLI are filled with observations from KCON. Additional data include daily average surface temperature, snow depth, and 24-h liquid-equivalent precipitation and snowfall from a NOAA Cooperative Observer Program Network site at PHM (614 m MSL) and hourly snow water equivalent (SWE) measurements from a National Resources Conservation Service Soil Climate Analysis Network snow-pillow site located in the Hubbard Brook Experimental Forecast (HB; 451 m MSL) (USDA 2019). Measurements of river stage height every 15 min at three USGS stream gauges on the Baker River in Rumney, Pemigewasset River in Woodstock, and Pemigewasset River in Plymouth (USGS 2019) provide insight into the hydrologic response of the rivers for the two case studies.

A portion of the case study and composite analyses are presented in a “time of maximum IVT” framework. This framework identifies the time of maximum IVT magnitude from the closest CFS grid point at 44.0°N, 71.5°W within the study region (Fig. 1c) during a 4-day period from 2 days before to 1 day after the day of the ice jam. Hydrometeorological data with a subdaily time scale are plotted relative to the time of maximum IVT (hereafter referred to as tmax) associated with each event, whereas hydrometeorological data with a daily time scale (e.g., snow depth observations at PHM) are plotted relative to the day of the ice jam in the case study and climatology analyses. Note that one ice jam (February 2017) contained two IVT magnitude maxima in the 2-day period before the ice jam and the latter, slightly weaker maxima closer in time to the date of the ice jam was chosen as tmax.

3. Case study analysis of two ice jams

This section describes the short-term antecedent and concurrent hydrometeorological characteristics of two locally high-impact ice jams on 26 February 2017 and 13 January 2018 for the period from 5 days before to 2 days after each event. A comparison of those similar and dissimilar short-term characteristics follows in section 3c with an additional analysis of longer-term antecedent and concurrent hydrometeorological characteristics for the period from 28 days before to 7 days after each event.

a. 26 February 2017 ice jam

The 26 February 2017 ice jam occurred in association with a maximum IVT magnitude of 367 kg m−1 s−1 at 0000 UTC 26 February 2017 (i.e., tmax). During the 5 days prior to tmax, daily maximum surface (2-m) temperatures at K1P1 exceeded 5°C and increased to ~20°C on 23 February 2017 (Fig. 2a). Surface dewpoint temperatures at both K1P1 and KMWN increased to >0°C for most of the 60-h period between 1800 UTC 23 February 2017 and 0600 UTC 26 February 2017 (Fig. 2a). The observations at lower-elevation K1P1 and upper-elevation KMWN suggest a warm and moist air mass over the study region for at least two days prior to tmax and decreasing temperatures at both sites following the passage of a cold front on the day of the ice jam (not shown). Total precipitation accumulation < 5 mm occurred at both K1P1 and KMWN during the 4.5-day period ending at 1200 UTC 25 February 2017 (i.e., 12 h before tmax), whereas total precipitation accumulation of 10–15 mm occurred during the 1-day period ending at 1200 UTC 26 February 2017 (i.e., 12 h after tmax; Fig. 2a). During the 2–4-day period with surface temperatures and dewpoint temperatures > 0°C at K1P1 and KMWN, the snow depth at PHM decreased 42% from ~110 cm at 1200 UTC 21 February 2017 to ~63 cm at 1200 UTC 26 February 2017 (Fig. 2b). Model-derived snow depth from the National Operational Hydrologic Remote Sensing Center (NOHRSC) illustrates a complete loss of snowpack in lower elevations (<500 m MSL) of the study region by 1200 UTC 25 February 2017 (not shown; data available via NOHRSC 2004). Total precipitation accumulation at PHM for the 25–27 February 2017 period, which occurred as rain, was 17 mm and total SWE loss at the HB snow-pillow site for the 21–27 February 2017 period was 86 mm (Fig. 2b).

Fig. 2.
Fig. 2.

The 7-day time series of surface observations and CFS-derived variables for the 26 Feb 2017 ice jam plotted relative to a tmax of 0000 UTC 26 Feb 2017. (a) Temperature (°C; solid line), dewpoint temperature (°C; dashed line), and 6-h liquid-equivalent precipitation accumulation (mm; vertical bars) at K1P1 (black/gray lines and black bars, respectively) and KMWN (red/pink lines and red bars, respectively). Six-hour liquid-equivalent precipitation accumulations are plotted at time values at the end of the 6-h period. (b) Snow depth (cm; blue line), 24-h snow accumulations (mm; gray bars), and 24-h liquid-equivalent precipitation (mm; green bars) at PHM and 24-h SWE loss (mm; orange bars) at HB. Data are plotted at 1200 UTC on each day and correspond to conditions over the previous 24 h. (c) CFS-derived instantaneous IWV (mm; gray line), IVT magnitude (kg m−1 s−1; black line) and precipitation rate [mm (6 h)−1; blue bars]. Precipitation rate values are plotted at the end of the 6-h period. (d) Hydrographs of stage height for gauges on the Pemigewasset River in Woodstock, NH (brown line); Baker River in Rumney, NH (green line); and Pemigewasset River in Plymouth, NH (black line). The horizontal black line at 13 ft denotes the flood stage at Plymouth, NH.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

Time series of CFS IWV and IVT magnitude at 44.0°N, 71.5°W illustrate an increase to local maxima in IWV values of ~25 mm and IVT magnitudes of ~370 kg m−1 s−1 at 1800 UTC 24 February 2017 prior to the slightly lower secondary maximum of 367 kg m−1 s−1 at tmax (Fig. 2a). The maximum CFS precipitation rate of ~11 mm (6 h)−1 for the 6-h period ending at 0600 UTC 26 February 2017 agrees with the observed precipitation rates and totals at K1P1 and KMWN for the same period. After tmax, the IWV value and IVT magnitude decrease, coincident with the arrival of drier air at K1P1 and KMWN behind the cold front (Fig. 2c). The river stage heights recorded at three stream gauges in the study region began rising on 24 February 2017 in response to temperature and dewpoint temperatures > 0°C and snowmelt into tributaries of the Pemigewasset River. A sharp increase in stage height occurred on the Pemigewasset River at Plymouth, New Hampshire, at ~0600 UTC 26 February 2017 indicating the formation of an ice jam downstream in Ashland, New Hampshire (Fig. 1b). In response to the formation of the ice jam, the Pemigewasset River crested at a stage height of ~15 ft, exceeding the local flood stage of 13 ft (Fig. 2d), and resulted in flooding of several areas in Plymouth and Holderness, New Hampshire (WMUR 2017).

The synoptic-scale flow at tmax contains a negatively tilted upper-tropospheric trough at 300 hPa and 500 hPa over the Great Lakes region associated with a pair of >35 m s−1 jet streaks at 300 hPa over the southeastern United States and eastern Canada and a ridge over the western North Atlantic and eastern Canada (Figs. 3a,b). The U.S. Northeast is located beneath the equatorward entrance region of the downstream anticyclonically curved upper-tropospheric jet streak, diffluent upper-tropospheric flow, and in a region of inferred cyclonic (positive) absolute vorticity advection by the geostrophic wind at 500 hPa that are all characteristic of synoptic-scale forcing for ascent (Figs. 3a,b). The lower-tropospheric circulation contains an 850-hPa and surface cyclone over the Hudson Bay and northern Quebec, with a cold front trailing south across the eastern United States and a warm front extending east into the Canadian Maritime provinces (Figs. 3c,d). The Northeast is located in the warm sector of this cyclone with 850-hPa temperatures of 6°–9°C under inferred lower-tropospheric warm air advection that is also characteristic of additional synoptic-scale forcing for ascent (Fig. 3c). Collocated with the warm sector of the cyclone is a narrow region of enhanced IWV > 20 mm and a corridor of IVT magnitudes > 250 kg m−1 s−1 along the prefrontal low-level jet that is characteristic of an AR along the U.S. East Coast (Fig. 3d).

Fig. 3.
Fig. 3.

Plan-view plots for CFS-derived variables at tmax of 0000 UTC 26 Feb 2017 of (a) 300-hPa geopotential height (dam; black contours), wind speed (m s−1; color fill plotted for values > 40 m s−1) and wind velocities (flag = 25 m s−1, barbs = 5 m s−1, half barbs = 2.5 m s−1), (b) 500-hPa geopotential height (dam; black contours), absolute vorticity (1 × 10−4 s−1; color fill for values > 16 × 10−4 s−1), and wind velocities as in (a) with the location of important trough and ridge axes discussed in text indicated by the dashed and solid black lines, respectively, (c) 850-hPa geopotential height (dam; black contours), temperature (°C; color fill), and wind velocities as in (a) with the locations of surface-based fronts are indicated in white, and (d) mean sea level pressure (hPa; black contours), integrated water vapor (mm; color shading for values > 20 mm), and integrated vapor transport (vectors; kg m−1 s−1; plotted for magnitudes > 250 kg m−1 s−1) with the locations of high- and low-pressure centers indicated by the blue “H” and red “L” symbols, respectively.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

b. 13 January 2018 ice jam

The 13 January 2018 ice jam occurred in association with an antecedent maximum IVT magnitude of 999 kg m−1 s−1 at 1800 UTC 12 January 2018 (i.e., tmax). During the 5 days prior to tmax, the surface temperature and dewpoint temperatures at both K1P1 and KMWN were predominantly <0°C with intermittent periods of light snow (Fig. 4a). Temperature and dewpoint temperatures increased to >0°C by 1800 UTC 11 January 2018 and remained >0°C for the 42-h period through 1200 UTC 13 January 2018. Maximum surface temperatures and dewpoint temperatures between 1800 UTC 11 January 2018 and 1200 UTC 13 January 2018 were ~5°–6°C with warmer temperatures observed at KMWN as compared to K1P1 associated with localized cold air damming in the region south and east of the White Mountains (not shown). Heavy precipitation [e.g., maximum precipitation rates > 15 mm (6 h)−1] occurred at K1P1 and KMWN between 1200 UTC 12 January 2018 and 0000 UTC 14 January 2018 with 36-h total precipitation accumulations of 42 and 56 mm, respectively. After 1200 UTC 13 January 2018, surface temperatures and dewpoint temperatures decreased below −15°C following the passage of a cold front (Fig. 4a). Prior to the ice jam, the snow depth at PHM increased to a maximum depth of ~58 cm on 1200 UTC 10 January 2018. The snow depth subsequently decreased 86% to ~8 cm by 1200 UTC 14 January 2018. Total precipitation accumulation at PHM (mostly rain) for the 24-h period ending 1200 UTC 13 January 2018 was 86 mm, whereas total SWE loss at the HB snow-pillow site for the 8–14 January 2018 period was 53 mm (Fig. 4b).

Fig. 4.
Fig. 4.

As in Fig. 2, but for the 13 Jan 2018 ice jam event with a tmax of 1800 UTC 12 Jan 2018.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

Time series of CFS IWV and IVT magnitude over the study region illustrate an antecedent increase in IWV values and IVT magnitudes to ~13 mm and 317 kg m−1 s−1, respectively, at 1800 UTC 8 January 2018 that coincided with snow accumulation at PHM for the 24-h period ending at 1200 UTC 9 January 2018 (Figs. 4b,c). IWV values and IVT magnitudes subsequently increased to 34 mm and 999 kg m−1 s−1, respectively, at tmax at 1800 UTC 13 January 2018, immediately prior to the ice jam (Fig. 4c). The average CFS precipitation rate between 1200 UTC 12 January 2018 and 1800 UTC 13 January 2018 was >10 mm (6 h)−1 (Fig. 4c) and corresponded well to the observed precipitation at K1P1 and KMWN (Fig. 4a). In contrast to the 26 February 2017 event, which featured an extended period of snowmelt resulting in a steady increase in river stage height prior to tmax, the environment prior to the 13 January 2018 event featured very little increase in stage height prior to tmax (Fig. 4d). The lack of antecedent hydrologic response is likely due to watershed-wide temperatures remaining predominantly below 0°C prior to 1800 UTC 11 January 2018. The stage height on the Pemigewasset River in Plymouth, New Hampshire, increased 12.1 ft in the 6-h period ending at 0600 UTC 13 January 2018 (Fig. 4d) coincident with the formation of the ice jam.

The synoptic-scale flow at tmax contains a positively tilted upper-tropospheric trough over central North America, a negatively tilted upper-tropospheric shortwave trough over the U.S. Southeast and a ridge over the western North Atlantic (Figs. 5a,b). The upper-tropospheric flow is characterized by a 90 m s−1 jet streak at 300 hPa that extended across eastern North America (Figs. 5a,b). The Northeast is located beneath the equatorward entrance region of the upper-tropospheric jet streak, and in a region of inferred positive absolute vorticity advection by the geostrophic wind at 500 hPa that is characteristic of synoptic-scale forcing for ascent (Figs. 5a,b). The lower-tropospheric circulation contains a trough along the U.S. East Coast and a broad subtropical anticyclone over the North Atlantic at 850 hPa (Fig. 5c). Southwesterly flow of 30–40 m s−1 and temperatures of 6°–12°C at 850 hPa along the U.S. East Coast result in inferred lower-tropospheric warm air advection that is also characteristic of synoptic-scale forcing for ascent (Fig. 5c). At the surface, a cyclone along the central U.S. East Coast is located beneath the equatorward entrance region of the 300-hPa jet streak (Fig. 5d). The warm sector of this cyclone contains a region of enhanced IWV values > 35 mm and a corridor of IVT magnitudes ~1000 kg m−1 s−1 that is characteristic of a strong AR along the U.S. East Coast (Fig. 5d).

Fig. 5.
Fig. 5.

As in Fig. 3, but for the 13 Jan 2018 ice jam event with a tmax of 1800 UTC 12 Jan 2018.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

c. Comparison of antecedent conditions

The two ice jams share similar antecedent synoptic-scale characteristics and somewhat dissimilar surface temperature, precipitation, and hydrologic responses, but interestingly that both led to flooding. For example, the synoptic-scale flow patterns at the time of maximum IVT magnitude, both preceding the time of the ice jam events, for both events illustrate an amplified large-scale flow pattern favoring synoptic-scale forcing for ascent that also contain regions of enhanced IWV values and IVT magnitudes characteristic of ARs along the U.S. East Coast. The February 2017 event featured a longer duration period of surface temperature and dewpoint temperatures > 0°C (~60–100 h), more melting of the antecedent snowpack (47-cm snow depth decrease and 86-mm SWE loss), and a smaller amount of rain (~10–20 mm) prior to the formation of the ice jam. In contrast, the January 2018 event featured a shorter period of surface temperature and dewpoint temperatures > 0°C (~48 h), less melting of the antecedent snowpack (50-cm snow depth decrease and 53-mm SWE loss), and a larger amount of rain (~40–90 mm) prior to the formation of the ice jam.

A time series of daily average temperature and snow depth at PHM for the period 28 days before to 7 days after the date of each ice jam is shown in Fig. 6. Both events contain a period of relatively cold daily average temperatures approximately two weeks prior to the date of each ice jam that persists for several days; we speculate that these conditions allowed for sustenance or growth of the regional snowpack and river ice. For example, the snow depth doubled prior to each event during these cold periods as precipitation fell as snow with daily average temperatures < 0°C. For the February 2017 event, the cold period persisted for ~5 days from 10 through 14 February 2017 with daily average temperatures of −10°C (Fig. 6a). For the January 2018 event, the cold period persisted for ~6 days from 28 December 2017 through 2 January 2018 with daily average temperatures of −20°C (Fig. 6b). Closer to the day of each ice jam, both events contained a period of relatively warm daily average temperatures (i.e., a midwinter thaw) with daily average temperatures > 5°C that resulted in a decrease in snow depth likely related to compaction, ripening, and ablation of the antecedent snowpack. The decrease in snow depth was ~47 cm (~42%) over an 8-day period from 18 to 26 February 2017 prior to the February 2017 ice jam and ~58 cm (~86%) over a 5-day period from 10 to 14 January 2018 prior to the January 2018 ice jam (Fig. 6). Although it is not known what the ice thicknesses were on the river prior to each event, the ~10-ft increases in stage height on the Pemigewasset River (Figs. 2d and 4d) were likely more than sufficient to produce fracturing of riverine ice that led to the ice jams (i.e., if the ice was less than 3 ft thick, the increase in stage height was greater than the 1.5–3 times the thickness).

Fig. 6.
Fig. 6.

Time series of daily average temperature (°C; red line) and snow depth (cm; blue line) at PHM for the ice jams on (a) 26 Feb 2017 and (b) 12 Jan 2018. The time series begin 28 days before and ends 7 days after the date of each ice jam (vertical black line). The horizontal black line denotes a daily average temperature of 0°C. Missing data are not plotted.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

4. Climatology of ice jams 1981–2019

a. Summary of 20 ice jams

The 20 ice jams on the Pemigewasset River at Plymouth, New Hampshire, during the CFS reanalysis and operational analysis period are listed with key hydrometeorological data in Table 1. The maximum IVT magnitude at tmax prior to each ice jam ranged from 290 to 1141 kg m−1 s−1 with a composite average magnitude of 664 kg m−1 s−1. The daily snow depth at PHM prior to and during each event decreased on average 47% (28 cm) with a range from 4% to 100%. This range suggests that a large amount of snowpack ablation is not always necessary for ice jam formation. Note that there was no snow depth recorded on the day of the 27 January 1986 ice jam event (i.e., no antecedent snowpack) and this event is therefore not included in the average snow depth loss calculation. The lower-elevation locations of K1P1, KLCI, and KCON (recall that KLCI and KCON observations are used for observations prior to 2005 and for missing observations as described in section 2) observed composite average surface temperatures and dewpoint temperatures at tmax of +7.4°C and +6.9°C, respectively, whereas upper-elevation KMWN observed composite average surface temperatures and dewpoint temperatures at tmax of +2.7°C and +2.7°C. These temperatures and dewpoint temperatures indicate that the entire Pemigewasset River watershed likely experienced, on average, warm, moist, and above-freezing conditions at tmax prior to the ice jams. The mean areal 48-h precipitation accumulations for a domain encompassed by 43.5°–44.5°N and 72.0°–71.0°W for each event spanned from 0.5 to 76.2 mm with an average of 34.6 mm. This range suggests that large precipitation accumulations are also not always necessary for ice jam formation.

b. Composite analyses

Composite time series analysis at PHM from 28 days before through 7 days after the dates of the 20 ice jams provide an assessment of the approximate mean hydrometeorological conditions at middle elevations over the study region (Fig. 7). The snowpack begins with a mean depth of ~38 cm at day −28 and steadily increases to 43 cm at day −12 and to ~52 cm by day −7 through day −2 (Fig. 7a). The increase in snowpack is due to periodic accumulations of snow in an environment characterized by below-normal daily average temperatures between −10°C and −5°C (Figs. 7a,b). We hypothesize that this period of below-normal daily average temperatures ≪ 0°C would have also likely resulted in increased growth of river ice before the ice jams based on the accumulated freezing degree-day methodology by USACE (2002, 2004). The snow depth subsequently decreases on average from 52 to 35 cm (−17 cm; 33%)1 coincident with a 4-day period of above-normal daily average temperatures ending on day+1 and ~40 mm of liquid-equivalent precipitation on day−0. The mean daily average temperature of ~1.7°C and <10 mm of snow accumulation observed on day −0 indicates that the ~40 mm of liquid-equivalent precipitation on that day fell primarily as rain (Fig. 7).

Fig. 7.
Fig. 7.

Composite time series for PHM observations from 28 days before through 7 days after the date of ice jams of (a) snow depth (cm; blue line), composite mean daily average temperature (°C; solid red line with circles) and climatological daily average temperature (°C; dashed red line). Confidence levels for statistically significant departures from climatology of the daily average temperatures are denoted by different color circles: 90% (blue dots), 95% (yellow dots), and 99% (red dots), and (b) 24-h snow accumulations (mm; gray bars) and liquid-equivalent precipitation (mm; blue bars). The gray shading in (a) denotes 4-day periods used for anomaly calculations in Fig. 8.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

The periods of below-normal and above-normal daily average temperatures (highlighted in gray in Fig. 7a) serve as the averaging periods for time-mean composite analyses of 500-hPa geopotential height and 850-hPa temperature patterns (Fig. 8). A length of four days is chosen for compositing based on the warm period duration from day −2 to day +1. The period from day −12 to day −9 is used for the corresponding cold period in order to include the significant departures of the daily average temperatures from climatology at day −12. The antecedent cold period composite analyses illustrate a mean 500-hPa trough over eastern North America characterized by geopotential height anomalies ~40 m below normal and 850-hPa temperature anomalies ~3°C below normal (Figs. 8a,c). Neither anomaly pattern during the cold period differs significantly from climatology at the 95% confidence level. Statistically significant negative anomalies during the cold period are present during shorter-duration compositing periods (not shown) suggesting that significant large-scale features associated with the colder temperatures at PHM between day −12 and day −7 may occur on more transient time scales. A large-scale regime transition is apparent as the warm period composite analyses illustrate an expansive region of positive anomalies in both 500-hPa geopotential heights (e.g., anomalies > 160 m above normal) and 850-hPa temperatures (e.g., anomalies > 8°C above normal) over eastern North America that are statistically significant with respect to climatology at the 95% confidence level (Figs. 8b,d). New Hampshire is located in a region of average 850-hPa temperatures between −2° and 0°C that are from +6°C to +7°C warmer than climatology, suggesting that boundary layer temperatures in the study region are likely above freezing during the warm period (Fig. 8d) in the absence of a significant temperature inversion.

Fig. 8.
Fig. 8.

Time-mean composite analyses of NCEP–NCAR reanalysis-derived means and anomalies for the PHM-based cold (from 12 through 9 days before ice jam dates) and warm (from 2 days prior through 1 day after ice jam dates) periods shown in Fig. 7a of (a) cold period 500-hPa geopotential heights (contoured every 6 dam) and anomalies (m; color shading), (b) warm period 500-hPa geopotential heights and anomalies as in (a), (c) cold period 850-hPa temperatures (contoured every 2°C) and anomalies (°C; color shading), and (d) warm period 850-hPa temperatures and anomalies as in (c). Statistical significance at the 95% confidence level is denoted by the thick black line.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

Composite time series analyses of several hydrometeorological characteristics over the study region from 120 h before (i.e., tmax − 120 h) to 48 h after tmax (i.e., tmax + 48 h) illustrates the regime transition from the end of the cold period through the anomalous warm period (Fig. 9). From tmax − 120 h to tmax − 36 h, locations in the valley (e.g., K1P1, KLCI, and KCON) and White Mountains (e.g., KMWN) contain mean observed surface temperatures < 0°C and dewpoint temperatures less than −10°C (Fig. 9a), mean CFS IWV values ~10 mm and IVT magnitudes ~150 kg m−1 s−1 (Figs. 9b,c), and mean CFS precipitation rates < 2 mm (6 h)−1 (Fig. 9d). Mean observed surface temperatures and dewpoint temperatures at locations in the valley and White Mountains increase to >5°C and ~2.5°C, respectively, at tmax ± 3 h and subsequently decrease to <0°C by tmax + 24 h. The locations in the valley remain above freezing for ~48 h with mean maximum surface temperatures and dewpoint temperatures of 7.4° and 6.9°C, respectively (Table 1 and Fig. 9a). The locations in the White Mountains remain above freezing for ~8 h with mean maximum temperatures and dewpoint temperatures of 2.7° and 2.6°C (Table 1 and Fig. 9a). The CFS IWV values, IVT magnitudes, and precipitation rates increase to mean maximum values of 25.5 mm, 664 kg m−1 s−1, and 10.8 mm (6 h)−1, respectively, between tmax − 6 h and tmax (Table 1 and Figs. 9b–d).

Fig. 9.
Fig. 9.

Composite time series plotted from 120 h before through 48 h after tmax of (a) observations of mean valley temperatures (°C; solid black line) and dewpoint temperatures (°C; dashed gray line) and mountain temperatures (°C; solid red line) and dewpoint temperatures (°C; dashed pink line) with the valley composite based on observations from K1P1, KLCI, and KCON and the mountain composite based on observations from KMWN, (b) CFS-derived boxplot distributions of IWV (mm) with the median (thick center bar), 25th and 75th percentile (bottom and top of box, respectively), hinge values (T-bars), and outliers (open circles), and mean value (gray line), (c) CFS-derived boxplot distributions of IVT magnitude (kg m−1 s−1) with corresponding values denoted as in (b), and (d) CFS-derived composite mean precipitation rate [mm (6 h)−1; green bars].

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

The synoptic-scale composite flow at tmax shares remarkable similarities to both case studies and contains a negatively tilted upper-tropospheric trough over eastern North America and a ridge over the western North Atlantic at 300 hPa (Fig. 10a). The upper-tropospheric flow also contains a jet streak with average maximum wind speeds > 45 m s−1 over the southeastern United States and a second jet streak with maximum wind speeds > 55 m s−1 over eastern Canada (Fig. 10a). The configuration of the dual jet streaks illustrates an upper-tropospheric pattern that favors a region of enhanced synoptic-scale forcing for ascent located between the two jet streaks that is typically linked to heavy precipitation events along the U.S. East Coast (e.g., Uccellini and Kocin 1987). The synoptic-scale forcing for ascent is also illustrated by a region of inferred positive absolute vorticity advection by the geostrophic wind at 500 hPa and warm air advection at 850 hPa over the Northeast (Figs. 10b,c). At the surface, the composite illustrates a cyclone with sea level pressure < 1000 hPa over northern New York and an anticyclone with sea level pressure > 1024 hPa over the western North Atlantic (Fig. 10d). The sea level pressure couplet results in a corridor of poleward water vapor flux along the U.S. East Coast containing IWV values ~30 mm and IVT magnitudes ~700 kg m−1 s−1 (Fig. 10d) characteristic of a AR.

Fig. 10.
Fig. 10.

As in Fig. 3, except for the 20-event composite at tmax. Note that the color bars in (a), (b), and (d) are different than those in Figs. 3 and 5 in order to better illustrate the composite mean values.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

c. Case-to-case variability

The climatology of 20 ice jams illustrates that all events were preceded by IVT magnitudes commonly associated with ARs (i.e., IVT magnitude > 250 kg m−1 s−1; Rutz et al. 2014). Analysis of each individual event at the time of maximum IVT magnitude, however, identifies that 19 of 20 events contain synoptic-scale patterns actually resembling ARs (i.e., long and narrow corridors of enhanced moisture transport > 2000 km in length and <1000 km in width in the warm sector of a midlatitude cyclone; Fig. 11). The one event not fitting this definition occurred on 30 March 1993 (Fig. 11n); however, that event followed a particularly extreme snowstorm earlier in the month known as the “Superstorm of 1993” (Bosart et al. 1996) that increased the snow depth at PHM to 119 cm on 14 March 1993. The 2-week period leading up to the outlier event on 30 March 1993 ice jam event had a daily average snow depth of 88 cm at PHM. This snow depth was 40% higher than the second largest 2-week snow depth of the 20 events studied (1996; not shown). This analysis suggests that midwinter regime transitions in association with environments characterized by ARs along the U.S. East Coast may provide the necessary synoptic-scale ingredients for ice jams and flooding over the Northeast given appropriate antecedent conditions (e.g., snow to melt and sufficient river ice). In these cases, the AR is representative of a synoptic-scale flow pattern that contains poleward transport of heat and moisture into the Northeast that may serve to modify and melt the preexisting snowpack in association with large sensible and latent heat fluxes (i.e., condensation fluxes). Previous studies indicate these fluxes often occur with near-surface dewpoint temperatures > 0°C and near-surface winds at valley elevations 5–10 m s−1 and at mountain elevations > 20 m s−1 (e.g., Moore and Owens 1984; Leathers et al. 1998). Note that the two case studies in 2017 and 2018 featured 60 and 42 h, respectively, of dewpoint temperatures > 0°C at both K1P1 and KMWN capable of producing a positive condensation flux into the snowpack.

Fig. 11.
Fig. 11.

Plan-view plot of CFSR-derived IVT magnitude (kg m−1 s−1; color shading) with vectors overlaid for magnitudes > 250 kg m−1 s−1 for the 20 ice jam events at each tmax. The grid point 44.0°N, 71.5°W is indicated by the blue dot in each panel.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

The AR may also represent a synoptic-scale flow pattern that favors rain and rain-on-snow due to the anomalous warmth and elevated freezing levels (Guan et al. 2016). Since ARs are characterized by a majority (~75%) of their water vapor flux below 2.25 km (Ralph et al. 2005) along a prefrontal low-level jet stream, they may also favor orographic enhanced precipitation as the low-level jet stream intersects the elevated terrain of the western White Mountains at the headwaters of the Pemigewasset watershed (e.g., ~1.0–1.6 km MSL). A 48-h composite mean precipitation analysis illustrates widespread precipitation > 25 mm across the U.S. Northeast with precipitation > 40–50 mm across the White Mountains in New Hampshire and northern Appalachians in Maine (Fig. 12a). The average orographic precipitation gradient is ~22 mm km−1 over the western White Mountains region (i.e., within the black box on Fig. 12a), with large variability across all 20 events (Fig. 12b). For example, 48-h precipitation for the 26 February 2017 event is ~4–12 mm at all elevations, whereas 48-h precipitation for the 13 January 2018 event ranged from 48 mm at 300–400 m MSL to 77 mm at 1200–1300 m MSL (Fig. 12b). Note that two other ice jam events on 24 December 1991 and 18 March 1990, in addition to 26 February 2017, also featured low precipitation (i.e., less than ~15 mm; Table 1) with small orographic gradients (Fig. 12b). Both the 26 February 2017 and 18 March 1990 events were characterized by 60–100 h of observed surface temperatures and dewpoint temperatures > 0°C and 36–48 h of IVT magnitudes > 250 kg m−1 s−1 that resulted in a large decrease of the snowpack prior to the respective ice jams. During these two ice jam events, it appears the lack of precipitation was offset by the extended period of snowmelt as compared to the other events which featured larger precipitation amounts (Table 1) and shorter melting periods (not shown). For the 24 December 1991 event, lower-elevation surface temperatures exceeded 0°C for only 24 h, however, this event was preceded by a similar short-duration period of warm temperatures associated with a prior AR on 21 December 1991 (not shown). For the 24 December 1991 ice jam, it appears that the lack of precipitation in the days immediately surrounding the event was therefore offset by an earlier period of antecedent precipitation and snowmelt, that may have combined to produce the hydrological response necessary for an ice jam.

Fig. 12.
Fig. 12.

PRISM-derived (a) composite mean 48-h precipitation (mm; shaded) for all 20 ice jams with the region used to calculate mean areal precipitation in Table 1 and the orographic gradients in (b) indicated by the white-dashed box, and (b) composite mean 48-h elevation-band-averaged precipitation (mm; based on 100-m elevation bins) with accumulations for the January 2018 (green line) and February 2017 (orange line) events and the 20-event composite average (black line) highlighted. Two low-precipitation events in March 1990 and December 1991 discussed in the text are also labeled. The 48-h period is determined as the 48 h ending at 1200 UTC on the calendar day of the ice jam.

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

5. Concluding discussion

Two ice jams on the Pemigewasset River in central New Hampshire in 2017 and 2018 that resulted in localized flooding occurred in association with different antecedent hydrometeorological conditions and similar synoptic-scale environments featuring ARs. The February 2017 event is described as a “long melting period with low precipitation” scenario, whereas the January 2018 event is described as a “short melting period with high precipitation” scenario. The February 2017 event is characterized by several days of maximum surface temperatures of 5°–20°C across the lower-elevation locations in New Hampshire and dewpoint temperatures > 0°C at the summit of Mount Washington that favored snowmelt and a steady rise in stage heights on the Pemigewasset River at Plymouth prior to the ice jam. Precipitation associated with synoptic-scale forcing for ascent in the presence of a warm, moist air mass and enhanced poleward water vapor transport (e.g., a weak AR with IVT magnitudes ~250–500 kg m−1 s−1) provided the final hydrologic trigger that increased river discharge and stage height to sufficient levels so as to break up, transport, and lodge river ice downstream of Plymouth into an ice jam. In contrast, the January 2018 event featured very little melting of the snowpack prior to the ice jam in association with surface temperatures and dewpoint temperatures remaining predominantly <0°C that resulted in minimal change in stage heights on the Pemigewasset River at Plymouth. A longer and more intense period of precipitation associated with synoptic-scale forcing for ascent in the presence of a warm, moist air mass and enhanced poleward water vapor transport (e.g., a strong AR with IVT magnitudes ~1000 kg m−1 s−1) resulted in a more rapid hydrologic response that increased river discharge and stage height to sufficient levels so as to break up, transport, and lodge river ice downstream of Plymouth into an ice jam.

The synoptic-scale environments of described by enhanced poleward water vapor transport, illustrated via enhanced IVT along ARs, in each case study is remarkably consistent with a composite of 20 ice jam events occurring on the Pemigewasset River during 1981–2019. A schematic summarizes the common synoptic-scale characteristics of the majority of the 20 ice jams (Fig. 13). A highly amplified upper-tropospheric flow pattern is associated with anomalously warm air over the U.S. Northeast and Maritime provinces of Canada that has the potential to modify the preexisting snowpack over the region. A negatively tilted upper-tropospheric trough over east-central North America, anticyclonically curved 300-hPa jet streak with maximum wind speeds > 55 m s−1 over eastern Canada, and concomitant synoptic-scale forcing for ascent results in an intensifying surface cyclone located in the Saint Lawrence River Valley. This surface cyclone is accompanied by an AR that describes a warm, humid air mass containing enhanced water vapor (mean IWV values > 25 mm) and enhanced poleward water vapor transport (mean IVT magnitudes > 600 kg m−1 s−1) that can lead to orographic precipitation, rain-on-snow, and snowmelt over the Pemigewasset watershed that culminates in an ice jam and possible flooding at Plymouth, New Hampshire. In this study, the AR perspective provides a framework for describing the synoptic-scale environment conducive to heavy rain and enhanced streamflow that may improve situational awareness in the ice jam and flood forecast process depending on antecedent conditions.

Fig. 13.
Fig. 13.

Overview schematic for meteorological features associated with ice jams on the Pemigewasset River in Plymouth, New Hampshire. The schematic highlights 1) the location of Plymouth (yellow star), 2) upper-level (e.g., 300-hPa) jet streaks (“J”; purple shading with dashed contours), 3) upstream trough in midlevel (e.g., 500-hPa) geopotential height contours (solid black lines), 4) warm-period temperature anomaly (red oval and shading), 5) tropospheric moisture (green shading for IWV values > 25 mm) and moisture transport (black arrow representing AR and corridor of IVT values > 600 kg m−1 s−1), 6) surface frontal features, and 7) surface cyclone and anticyclone locations (red “L” and blue “H” symbols, respectively) and associated lower-tropospheric flow (red arrows).

Citation: Journal of Hydrometeorology 21, 12; 10.1175/JHM-D-20-0027.1

As evidenced by two case studies and a unique composite analysis of 20 ice jam events, this study identified that ice jams and associated flooding in central New Hampshire can be summarized into five general phases that may improve situational awareness:

  1. 1–2 weeks before the ice jam: Typical winter conditions associated with normal or slightly below-normal temperatures and periodic accumulations of snow serve to increase river ice and build the regional snowpack, respectively.

  2. 2–4 days before the ice jam: A midwinter thaw associated with above-normal temperatures modifies the snowpack. Rivers begins to rise as snowmelt enters the stream network. River ice may begin to breakup against the increased streamflow.

  3. 0–1 days before the ice jam: A midlatitude cyclone located over the Saint Lawrence River Valley containing water vapor and water vapor transport characteristics of an AR transport warm, humid air poleward along the U.S. East Coast. A period of heavy precipitation occurs coincident with enhanced IWV and IVT magnitudes over the region (e.g., IWV > 25 mm and IVT magnitudes > 600 kg m−1 s−1). Streams in the watershed swell and river ice begins to flow downstream.

  4. Day of the ice jam: Ice floes aggregate at a constriction, bend, or obstruction in the river. The river channel becomes completely blocked, and a sharp increase in stage height occurs upstream of the jam, possibly resulting in flooding of the local area.

  5. After the ice jam: Depending on the synoptic-scale conditions, the jam and associated flooding may freeze in place if a cold front passes and temperatures drop below freezing for an extended period

The differences in the antecedent hydrometeorological conditions described for the 2017 and 2018 events is also characteristic of the case-to-case variability of the 20 ice jam events with respect to differences in precipitation and snow depth that make each case unique. As such, the length of time in each phase mentioned above will likely differ from event to event, some phases may overlap, and there is likely a spectrum of possible snowmelt and precipitation combinations based on observed variability.

While the 20 events investigated in this study are sufficient in order to gain a novel understanding of the similarities in the large-scale and environmental characteristics associated with ice jams and flooding in the Northeast, the case-to-case variability in antecedent hydrometeorological conditions necessitate additional analyses to better understand how all these different ingredients may combine to increase our ability to better predict ice jam formation. Similarly, the extensive record contained within the Ice Jam Database provides an opportunity to expand this climatology to other locations (e.g., Fig. 1a) and also further back in time. The number and variety of watersheds where ice jams occur suggests that small differences in the antecedent hydrometeorological condition or small differences in temperature, moisture, and wind speed/direction may lead to an ice jam forming in one watershed but not an adjacent one during the same synoptic-scale event. These localized differences, and likely incomplete historical record of ice jam events, likely create challenges in using objective criteria to successfully predict ice jam events with a low false alarm ratio; however, the methods provided in this study may provide useful situational awareness in the prediction of future events. Additional work is also aimed at investigating the role of sensible and latent heat fluxes and its influence on modifying and melting the snowpack, especially when they occur in association with enhanced wind speeds and moisture during environments characterized by ARs along the U.S. East Coast.

Acknowledgments

This research was supported by the Plymouth State University Center for the Environment. The authors thank Dr. Eric Hoffman (Plymouth State University) who provided valuable feedback during the completion of a M.S. thesis by the first author at Plymouth State University that contains a portion of the results presented in this manuscript, and two anonymous reviewers whose feedback improved the quality of this manuscript.

REFERENCES

  • Agel, L., M. Barlow, F. Colby, H. Binder, J. L. Catto, A. Hoell, and J. Cohen, 2019: Dynamical analysis of extreme precipitation in the us northeast based on large-scale meteorological patterns. Climate Dyn., 52, 17391760, https://doi.org/10.1007/s00382-018-4223-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barlow, M., and Coauthors, 2019: North American extreme precipitation events and related large-scale meteorological patterns: A review of statistical methods, dynamics, modeling, and trends. Climate Dyn., 53, 68356875, https://doi.org/10.1007/s00382-019-04958-z.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bates, R., and M.-L. Brown, 1982: Meteorological conditions causing major ice jam formation and flooding on the Ottauquechee River, Vermont. CRREL Special Rep. 826, 28 pp., http://hdl.handle.net/11681/12287.

  • Beltaos, S., and J. Wong, 1986: Preliminary studies of grounded ice jams. IAHR Ice Symp., Iowa City, IA, International Association for Hydro-Environment Engineering and Research, 112, http://publications.gc.ca/collections/collection_2018/eccc/En13-5-86-108-eng.pdf.

    • Search Google Scholar
    • Export Citation
  • Bennett, K. E., and T. D. Prowse, 2010: Northern Hemisphere geography of ice-covered rivers. Hydrol. Processes, 24, 235240, https://doi.org/10.1002/hyp.7561.

    • Search Google Scholar
    • Export Citation
  • Bieniek, P. A., U. S. Bhatt, L. A. Rundquist, S. D. Lindsey, X. Zhang, and R. L. Thoman, 2011: Large-scale climate controls of interior Alaska River ice breakup. J. Climate, 24, 286297, https://doi.org/10.1175/2010JCLI3809.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosart, L. F., G. J. Hakim, K. R. Tyle, M. A. Bedrick, W. E. Bracken, M. J. Dickinson, and D. M. Schultz, 1996: Large-scale antecedent conditions associated with the 12–14 March 1993 cyclone (“Superstorm ’93”) over eastern North America. Mon. Wea. Rev., 124, 18651891, https://doi.org/10.1175/1520-0493(1996)124<1865:LSACAW>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosilovich, M. G., J. Chen, F. R. Robertson, and R. F. Adler, 2008: Evaluation of global precipitation in reanalyses. J. Appl. Meteor. Climatol., 47, 22792299, https://doi.org/10.1175/2008JAMC1921.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cordeira, J. M., F. M. Ralph, A. Martin, N. Gaggini, J. R. Spackman, P. J. Neiman, J. Rutz, and R. Pierce, 2017: Forecasting atmospheric rivers during CalWater 2015. Bull. Amer. Meteor. Soc., 98, 449459, https://doi.org/10.1175/BAMS-D-15-00245.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Daly, C., R. P. Neilson, and D. L. Phillips, 1994: A statistical-topographical model for mapping climatological precipitation over mountainous terrain. J. Appl. Meteor., 33, 140158, https://doi.org/10.1175/1520-0450(1994)033<0140:ASTMFM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dettinger, M. D., 2013: Atmospheric rivers as drought busters on the U.S. West Coast. J. Hydrometeor., 14, 17211732, https://doi.org/10.1175/JHM-D-13-02.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dettinger, M. D., F. M. Ralph, T. Das, P. J. Neiman, and D. R. Cayan, 2011: Atmospheric rivers, floods and the water resources of California. Water, 3, 445478, https://doi.org/10.3390/w3020445.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • FEMA, 2018: Guides for flood risk analysis and mapping: Ice-jam analysis and mapping. FEMA Guidance Doc. 94, 19 pp., https://www.fema.gov/media-library-data/1520964160270-7c49e1753d0b2634e0c5fb4999459374/Ice_Jam_Guidance_Feb_2018.pdf.

  • FloodList, 2018: USA and Canada – 100s evacuated as ice jams cause floding. 17 January, accessed 8 July 2020, http://floodlist.com/america/evacuations-ice-jams-flooding-north-east-january-2018.

  • French, H. M., 2018: The Periglacial Environment. Wiley, 515 pp.

  • Graybeal, D. Y., and D. J. Leathers, 2006: Snowmelt-related flood risk in Appalachia: First estimates from a historical snow climatology. J. Appl. Meteor. Climatol., 45, 178193, https://doi.org/10.1175/JAM2330.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guan, B., N. P. Molotch, D. E. Waliser, E. J. Fetzer, and P. J. Neiman, 2010: Extreme snowfall events linked to atmospheric rivers and surface air temperature via satellite measurements. Geophys. Res. Lett., 37, L20401, https://doi.org/10.1029/2010GL044696.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guan, B., D. E. Waliser, F. M. Ralph, E. J. Fetzer, and P. J. Neiman, 2016: Hydrometeorological characteristics of rain-on-snow events associated with atmospheric rivers. Geophys. Res. Lett., 43, 29642973, https://doi.org/10.1002/2016GL067978.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hart, R. H., and R. E. Grumm, 2001: Using normalized climatological anomalies to rank synoptic-scale events objectively. Mon. Wea. Rev., 129, 24262442, https://doi.org/10.1175/1520-0493(2001)129<2426:UNCATR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hecht, C. W., and J. M. Cordeira, 2017: Characterizing the influence of atmospheric river orientation and intensity on precipitation distributions over North Coastal California. Geophys. Res. Lett., 44, 90489058, https://doi.org/10.1002/2017GL074179.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, https://doi.org/10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lamjiri, M. A., M. D. Dettinger, F. M. Ralph, and B. Guan, 2017: Hourly storm characteristics along the U.S. West Coast: Role of atmospheric rivers in extreme precipitation. Geophys. Res. Lett., 44, 70207028, https://doi.org/10.1002/2017GL074193.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lapenta, K. D., and Coauthors, 1995: The challenge of forecasting heavy rain and flooding throughout the eastern region of the national weather Service. Part I: Characteristics and events. Wea. Forecasting, 10, 7890, https://doi.org/10.1175/1520-0434(1995)010<0078:TCOFHR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Leathers, D. J., D. R. Kluck, and S. Kroczynski, 1998: The severe flooding event of January 1996 across north-central Pennsylvania. Bull. Amer. Meteor. Soc., 79, 785797, https://doi.org/10.1175/1520-0477(1998)079<0785:TSFEOJ>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Madaeni, F., R. Lhissou, K. Chokmani, S. Raymond, and Y. Gauthier, 2020: Ice jam formation, breakup and prediction methods based on hydroclimatic data using artificial intelligence: A review. Cold Reg. Sci. Technol., 174, 103032, https://doi.org/10.1016/j.coldregions.2020.103032.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., C. F. Chappell, and L. R. Hoxit, 1979: Synoptic and Meso-α scale aspects of flash flood events. Bull. Amer. Meteor. Soc., 60, 115123, https://doi.org/10.1175/1520-0477-60.2.115.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Magnuson, J. J., and Coauthors, 2000: Historical trends in lake and river ice cover in the Northern Hemisphere. Science, 289, 17431746, https://doi.org/10.1126/science.289.5485.1743.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahoney, K. D. L., and Coauthors, 2016: Understanding the role of atmospheric rivers in heavy precipitation in the southeast United States. Mon. Wea. Rev., 144, 16171632, https://doi.org/10.1175/MWR-D-15-0279.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miller, D. K., D. Holtz, J. Winton, and L. Stewart, 2018: Investigation of atmospheric rivers impacting the Pigeon River basin of the southern Appalachian Mountains. Wea. Forecasting, 33, 283299, https://doi.org/10.1175/WAF-D-17-0060.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, B. J., P. J. Neiman, F. M. Ralph, and F. E. Barthold, 2012: Physical processes associated with heavy flooding rainfall in Nashville, Tennessee, and vicinity during 1–2 May 2010: The role of an atmospheric river and mesoscale convective systems. Mon. Wea. Rev., 140, 358378, https://doi.org/10.1175/MWR-D-11-00126.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, R. D., and I. F. Owens, 1984: Controls on advective snowmelt in a maritime alpine basin. J. Climate Appl. Meteor., 23, 135142, https://doi.org/10.1175/1520-0450(1984)023<0135:COASIA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • NCEI, 2019: Climate Data Online. Accessed 25 February 2019, https://www.ncdc.noaa.gov/cdo-web/.

  • Neiman, P. J., F. M. Ralph, A. B. White, D. E. Kingsmill, and P. O. G. Persson, 2002: The statistical relationship between upslope flow and rainfall in California’s Coastal Mountains: Observations during CALJET. Mon. Wea. Rev., 130, 14681492, https://doi.org/10.1175/1520-0493(2002)130<1468:TSRBUF>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., F. M. Ralph, G. A. Wick, J. D. Lundquist, and M. D. Dettinger, 2008: Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting the West Coast of North America based on eight years of SSM/I satellite observations. J. Hydrometeor., 9, 2247, https://doi.org/10.1175/2007JHM855.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Newell, R. E., N. E. Newell, Y. Zhu, and C. Scott, 1992: Tropospheric rivers?—A pilot study. Geophys. Res. Lett., 19, 24012404, https://doi.org/10.1029/92GL02916.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • NOHRSC, 2004: Snow Data Assimilation System (SNODAS) data products at NSIDC. National Snow and Ice Data Center, accessed 1 April 2019, https://www.nohrsc.noaa.gov/archived_data/.

  • Palecki, M. A., and R. G. Barry, 1986: Freeze-up and break-up of lakes as an index of temperature changes during the transition seasons: A case study for Finland. J. Climate Appl. Meteor., 25, 893902, https://doi.org/10.1175/1520-0450(1986)025<0893:FUABUO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pradhanang, S. M., A. Frei, M. Zion, E. M. Schneiderman, T. S. Steenhuis, and D. Pierson, 2013: Rain-on-snow-runoff events in New York. Hydrol. Processes, 27, 30353049, https://doi.org/10.1002/hyp.9864.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph, F. M., P. J. Neiman, and R. Rotunno, 2005: Dropsonde observations in low-level jets over the northeastern Pacific Ocean from CALJET-1998 and PACJET-2001: Mean vertical-profile and atmospheric-river characteristics. Mon. Wea. Rev., 133, 889910, https://doi.org/10.1175/MWR2896.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph, F. M., P. J. Neiman, G. A. Wick, S. I. Gutman, M. D. Dettinger, D. R. Cayan, and A. B. White, 2006: Flooding on California’s Russian River: Role of atmospheric rivers. Geophys. Res. Lett., 33, L13801, https://doi.org/10.1029/2006GL026689.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph, F. M., J. M. Cordeira, P. J. Neiman, and M. Hughes, 2016: Landfalling atmospheric rivers, the Sierra barrier jet, and extreme daily precipitation in Northern California’s Upper Sacramento River watershed. J. Hydrometeor., 17, 19051914, https://doi.org/10.1175/JHM-D-15-0167.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph, F. M., J. J. Rutz, J. M. Cordeira, M. Dettinger, M. Anderson, D. Reynolds, L. J. Schick, and C. Smallcomb, 2019: A scale to characterize the strength and impacts of atmospheric rivers. Bull. Amer. Meteor. Soc., 100, 269289, https://doi.org/10.1175/BAMS-D-18-0023.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph, R. M., P. J. Neiman, and G. A. Wick, 2004: Satellite and CALJET aircraft observations of atmospheric rivers over the eastern North Pacific Ocean during the winter of 1997/98. Mon. Wea. Rev., 132, 17211745, https://doi.org/10.1175/1520-0493(2004)132<1721:SACAOO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ramos, A. M., R. M. Trigo, M. L. R. Liberato, and R. Tomé, 2015: Daily precipitation extreme events in the Iberian Peninsula and its association with atmospheric rivers. J. Hydrometeor., 16, 579597, https://doi.org/10.1175/JHM-D-14-0103.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Robertson, D. M., R. A. Ragotzkie, and J. J. Magnuson, 1992: Lake ice records used to detect historical and future climate changes. Climatic Change, 21, 407427, https://doi.org/10.1007/BF00141379.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rokaya, P., S. Budhathoki, and K.-E. Lindenschmidt, 2018: Ice-jam flood research: A scoping review. Nat. Hazards, 94, 14391457, https://doi.org/10.1007/s11069-018-3455-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Romolo, L., T. D. Prowse, D. Blair, B. R. Bonsal, and L. W. Martz, 2006a: The synoptic climate controls on hydrology in the upper reaches of the Peace River Basin. Part I: Snow accumulation. Hydrol. Processes, 20, 40974111, https://doi.org/10.1002/hyp.6421.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Romolo, L., T. D. Prowse, D. Blair, B. R. Bonsal, and L. W. Martz, 2006b: The synoptic climate controls on hydrology in the upper reaches of the Peace River Basin. Part II: Snow ablation. Hydrol. Processes, 20, 41134129, https://doi.org/10.1002/hyp.6422.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rutz, J. J., W. J. Steenburgh, and F. M. Ralph, 2014: Climatological characteristics of atmospheric rivers and their inland penetration over the western United States. Mon. Wea. Rev., 142, 905921, https://doi.org/10.1175/MWR-D-13-00168.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saha, S., and Coauthors, 2006: The NCEP Climate Forecast System. J. Climate, 19, 34833517, https://doi.org/10.1175/JCLI3812.1.

  • Saha, S., and Coauthors, 2014: The NCEP Climate Forecast System version 2. J. Climate, 27, 21852208, https://doi.org/10.1175/JCLI-D-12-00823.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shields, C. A., and Coauthors, 2018: Atmospheric River Tracking Method Intercomparison Project (ARTMIP): Project goals and experimental design. Geosci. Model Dev., 11, 24552474, https://doi.org/10.5194/gmd-11-2455-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, B. L., S. E. Yuter, P. J. Neiman, and D. E. Kingsmill, 2010: Water vapor fluxes and orographic precipitation, over Northern California associated with a landfalling atmospheric river. Mon. Wea. Rev., 138, 74100, https://doi.org/10.1175/2009MWR2939.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Town of Plymouth New Hampshire, 2016: Town of Plymouth New Hampshire hazard mitigation plan. 68 pp., http://www.plymouth-nh.org/wp-content/uploads/2014/04/PlymouthNH_HMP_ADOPTED9.26.16.pdf.

  • UCAR, 2020: River ice processes. The Comet Program, accessed 8 July 2020, https://www.meted.ucar.edu/hydro/basic/RiverIce/index.htm.

  • Uccellini, L. W., and P. J. Kocin, 1987: The interaction of jet streak circulations during heavy snow events along the East Coast of the United States. Wea. Forecasting, 2, 289308, https://doi.org/10.1175/1520-0434(1987)002<0289:TIOJSC>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • USACE, 2002: Engineering and design: Ice engineering. U.S. Army Corps of Engineers Manual 1110-2-1612, 475 pp., https://www.nrc.gov/docs/ML0901/ML090150033.pdf.

  • USACE, 2004: Ice engineering: Method to estimate river ice thickness based on meteorological data. ERDC/CRREL Tech. Note 04-3, 7 pp., https://apps.dtic.mil/dtic/tr/fulltext/u2/a430678.pdf.

  • USACE, 2019: Ice Jam Database. Accessed 13 March 2019, https://icejam.sec.usace.army.mil/.

  • USDA, 2019: Hubbard Brook (2069) – Site information and reports. Accessed 10 April 2019, https://wcc.sc.egov.usda.gov/nwcc/site?sitenum=2069.

  • USGS, 2019: Water data for the nation. Accessed 25 February 2019, https://waterdata.usgs.gov/nwis.

  • Weyrick, P. B., K. D. White, S. F. Daly, M. J. Bullock, and J. J. Gagnon, 2007: CRREL’s ice jam database and website. 14th Workshop on the Hydraulics of Ice Covered Rivers, Quebec City, QC, Canada, Canadian Geophysical Union Hydrology Section, 14 pp., http://www.cripe.ca/docs/proceedings/14/Weyrick-et-al-2007.pdf.

    • Search Google Scholar
    • Export Citation
  • White, A. B., B. J. Moore, D. J. Gottas, and P. J. Neiman, 2019: Winter storm conditions leading to excessive runoff above California’s Oroville Dam during January and February 2017. Bull. Amer. Meteor. Soc., 100, 5570, https://doi.org/10.1175/BAMS-D-18-0091.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • White, K. D., and H. J. Eames, 1999: CRREL Ice Jam Database. CRREL Rep. 992, 24 pp., http://hdl.handle.net/11681/9250.

  • Wilks, D. S., 2006: Statistical Methods in the Atmospheric Sciences. 2nd ed. Academic Press, 627 pp.

  • WMUR, 2017: 4-mile ice jam causes flooding near Plymouth State University. 27 February, accessed 12 March 2019, https://www.wmur.com/article/exit-25-on-i-93-route-175a-closed-for-flooding-due-to-ice-jam/8982054.

  • Zufelt, J. E., and M. A. Bilello, 1992: Effects of severe freezing periods and discharge on the formation of ice jams at Salmon, Idaho. CRREL Rep. 9214, 19 pp., http://hdl.handle.net/11681/9148.

1

The snow depth decrease described here is different than the decrease described in Table 1 due to the different methods of calculation used for each average. For the table-based calculation the snow depth decrease associated with each ice jam event was calculated first, then averaged over all events to obtain the 28 cm (47%) decrease. For the time-series-based calculation the average snow depth for each day relative to the composite ice jam date was calculated first, followed by determining the overall snow depth decrease from the mean to obtain the 17 cm (33%) decrease. From the perspective of this study the average decrease provided in Table 1 is considered more representative of the real-environment conditions.

Save
  • Agel, L., M. Barlow, F. Colby, H. Binder, J. L. Catto, A. Hoell, and J. Cohen, 2019: Dynamical analysis of extreme precipitation in the us northeast based on large-scale meteorological patterns. Climate Dyn., 52, 17391760, https://doi.org/10.1007/s00382-018-4223-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barlow, M., and Coauthors, 2019: North American extreme precipitation events and related large-scale meteorological patterns: A review of statistical methods, dynamics, modeling, and trends. Climate Dyn., 53, 68356875, https://doi.org/10.1007/s00382-019-04958-z.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bates, R., and M.-L. Brown, 1982: Meteorological conditions causing major ice jam formation and flooding on the Ottauquechee River, Vermont. CRREL Special Rep. 826, 28 pp., http://hdl.handle.net/11681/12287.

  • Beltaos, S., and J. Wong, 1986: Preliminary studies of grounded ice jams. IAHR Ice Symp., Iowa City, IA, International Association for Hydro-Environment Engineering and Research, 112, http://publications.gc.ca/collections/collection_2018/eccc/En13-5-86-108-eng.pdf.

    • Search Google Scholar
    • Export Citation
  • Bennett, K. E., and T. D. Prowse, 2010: Northern Hemisphere geography of ice-covered rivers. Hydrol. Processes, 24, 235240, https://doi.org/10.1002/hyp.7561.

    • Search Google Scholar
    • Export Citation
  • Bieniek, P. A., U. S. Bhatt, L. A. Rundquist, S. D. Lindsey, X. Zhang, and R. L. Thoman, 2011: Large-scale climate controls of interior Alaska River ice breakup. J. Climate, 24, 286297, https://doi.org/10.1175/2010JCLI3809.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosart, L. F., G. J. Hakim, K. R. Tyle, M. A. Bedrick, W. E. Bracken, M. J. Dickinson, and D. M. Schultz, 1996: Large-scale antecedent conditions associated with the 12–14 March 1993 cyclone (“Superstorm ’93”) over eastern North America. Mon. Wea. Rev., 124, 18651891, https://doi.org/10.1175/1520-0493(1996)124<1865:LSACAW>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosilovich, M. G., J. Chen, F. R. Robertson, and R. F. Adler, 2008: Evaluation of global precipitation in reanalyses. J. Appl. Meteor. Climatol., 47, 22792299, https://doi.org/10.1175/2008JAMC1921.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cordeira, J. M., F. M. Ralph, A. Martin, N. Gaggini, J. R. Spackman, P. J. Neiman, J. Rutz, and R. Pierce, 2017: Forecasting atmospheric rivers during CalWater 2015. Bull. Amer. Meteor. Soc., 98, 449459, https://doi.org/10.1175/BAMS-D-15-00245.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Daly, C., R. P. Neilson, and D. L. Phillips, 1994: A statistical-topographical model for mapping climatological precipitation over mountainous terrain. J. Appl. Meteor., 33, 140158, https://doi.org/10.1175/1520-0450(1994)033<0140:ASTMFM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dettinger, M. D., 2013: Atmospheric rivers as drought busters on the U.S. West Coast. J. Hydrometeor., 14, 17211732, https://doi.org/10.1175/JHM-D-13-02.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dettinger, M. D., F. M. Ralph, T. Das, P. J. Neiman, and D. R. Cayan, 2011: Atmospheric rivers, floods and the water resources of California. Water, 3, 445478, https://doi.org/10.3390/w3020445.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • FEMA, 2018: Guides for flood risk analysis and mapping: Ice-jam analysis and mapping. FEMA Guidance Doc. 94, 19 pp., https://www.fema.gov/media-library-data/1520964160270-7c49e1753d0b2634e0c5fb4999459374/Ice_Jam_Guidance_Feb_2018.pdf.

  • FloodList, 2018: USA and Canada – 100s evacuated as ice jams cause floding. 17 January, accessed 8 July 2020, http://floodlist.com/america/evacuations-ice-jams-flooding-north-east-january-2018.

  • French, H. M., 2018: The Periglacial Environment. Wiley, 515 pp.

  • Graybeal, D. Y., and D. J. Leathers, 2006: Snowmelt-related flood risk in Appalachia: First estimates from a historical snow climatology. J. Appl. Meteor. Climatol., 45, 178193, https://doi.org/10.1175/JAM2330.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guan, B., N. P. Molotch, D. E. Waliser, E. J. Fetzer, and P. J. Neiman, 2010: Extreme snowfall events linked to atmospheric rivers and surface air temperature via satellite measurements. Geophys. Res. Lett., 37, L20401, https://doi.org/10.1029/2010GL044696.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guan, B., D. E. Waliser, F. M. Ralph, E. J. Fetzer, and P. J. Neiman, 2016: Hydrometeorological characteristics of rain-on-snow events associated with atmospheric rivers. Geophys. Res. Lett., 43, 29642973, https://doi.org/10.1002/2016GL067978.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hart, R. H., and R. E. Grumm, 2001: Using normalized climatological anomalies to rank synoptic-scale events objectively. Mon. Wea. Rev., 129, 24262442, https://doi.org/10.1175/1520-0493(2001)129<2426:UNCATR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hecht, C. W., and J. M. Cordeira, 2017: Characterizing the influence of atmospheric river orientation and intensity on precipitation distributions over North Coastal California. Geophys. Res. Lett., 44, 90489058, https://doi.org/10.1002/2017GL074179.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, https://doi.org/10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lamjiri, M. A., M. D. Dettinger, F. M. Ralph, and B. Guan, 2017: Hourly storm characteristics along the U.S. West Coast: Role of atmospheric rivers in extreme precipitation. Geophys. Res. Lett., 44, 70207028, https://doi.org/10.1002/2017GL074193.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lapenta, K. D., and Coauthors, 1995: The challenge of forecasting heavy rain and flooding throughout the eastern region of the national weather Service. Part I: Characteristics and events. Wea. Forecasting, 10, 7890, https://doi.org/10.1175/1520-0434(1995)010<0078:TCOFHR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Leathers, D. J., D. R. Kluck, and S. Kroczynski, 1998: The severe flooding event of January 1996 across north-central Pennsylvania. Bull. Amer. Meteor. Soc., 79, 785797, https://doi.org/10.1175/1520-0477(1998)079<0785:TSFEOJ>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Madaeni, F., R. Lhissou, K. Chokmani, S. Raymond, and Y. Gauthier, 2020: Ice jam formation, breakup and prediction methods based on hydroclimatic data using artificial intelligence: A review. Cold Reg. Sci. Technol., 174, 103032, https://doi.org/10.1016/j.coldregions.2020.103032.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., C. F. Chappell, and L. R. Hoxit, 1979: Synoptic and Meso-α scale aspects of flash flood events. Bull. Amer. Meteor. Soc., 60, 115123, https://doi.org/10.1175/1520-0477-60.2.115.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Magnuson, J. J., and Coauthors, 2000: Historical trends in lake and river ice cover in the Northern Hemisphere. Science, 289, 17431746, https://doi.org/10.1126/science.289.5485.1743.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahoney, K. D. L., and Coauthors, 2016: Understanding the role of atmospheric rivers in heavy precipitation in the southeast United States. Mon. Wea. Rev., 144, 16171632, https://doi.org/10.1175/MWR-D-15-0279.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miller, D. K., D. Holtz, J. Winton, and L. Stewart, 2018: Investigation of atmospheric rivers impacting the Pigeon River basin of the southern Appalachian Mountains. Wea. Forecasting, 33, 283299, https://doi.org/10.1175/WAF-D-17-0060.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, B. J., P. J. Neiman, F. M. Ralph, and F. E. Barthold, 2012: Physical processes associated with heavy flooding rainfall in Nashville, Tennessee, and vicinity during 1–2 May 2010: The role of an atmospheric river and mesoscale convective systems. Mon. Wea. Rev., 140, 358378, https://doi.org/10.1175/MWR-D-11-00126.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, R. D., and I. F. Owens, 1984: Controls on advective snowmelt in a maritime alpine basin. J. Climate Appl. Meteor., 23, 135142, https://doi.org/10.1175/1520-0450(1984)023<0135:COASIA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • NCEI, 2019: Climate Data Online. Accessed 25 February 2019, https://www.ncdc.noaa.gov/cdo-web/.

  • Neiman, P. J., F. M. Ralph, A. B. White, D. E. Kingsmill, and P. O. G. Persson, 2002: The statistical relationship between upslope flow and rainfall in California’s Coastal Mountains: Observations during CALJET. Mon. Wea. Rev., 130, 14681492, https://doi.org/10.1175/1520-0493(2002)130<1468:TSRBUF>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., F. M. Ralph, G. A. Wick, J. D. Lundquist, and M. D. Dettinger, 2008: Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting the West Coast of North America based on eight years of SSM/I satellite observations. J. Hydrometeor., 9, 2247, https://doi.org/10.1175/2007JHM855.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Newell, R. E., N. E. Newell, Y. Zhu, and C. Scott, 1992: Tropospheric rivers?—A pilot study. Geophys. Res. Lett., 19, 24012404, https://doi.org/10.1029/92GL02916.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • NOHRSC, 2004: Snow Data Assimilation System (SNODAS) data products at NSIDC. National Snow and Ice Data Center, accessed 1 April 2019, https://www.nohrsc.noaa.gov/archived_data/.

  • Palecki, M. A., and R. G. Barry, 1986: Freeze-up and break-up of lakes as an index of temperature changes during the transition seasons: A case study for Finland. J. Climate Appl. Meteor., 25, 893902, https://doi.org/10.1175/1520-0450(1986)025<0893:FUABUO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pradhanang, S. M., A. Frei, M. Zion, E. M. Schneiderman, T. S. Steenhuis, and D. Pierson, 2013: Rain-on-snow-runoff events in New York. Hydrol. Processes, 27, 30353049, https://doi.org/10.1002/hyp.9864.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ralph,