Snow Level Characteristics and Impacts of a Spring 2 Typhoon-originating Atmospheric River in the Sierra 3 Nevada , USA 4

On 5-7 April 2018 a landfalling atmospheric river resulted in widespread heavy 10 precipitation in the Sierra Nevada of California and Nevada. Observed snow levels during this 11 event were among the highest snow levels recorded since observations began in 2002 and exceeded 12 2.75 km for 31 hours in the northern Sierra Nevada and 3.75 km for 12 hours in the southern Sierra 13 Nevada. The anomalously high snow levels and over 80 mm of precipitation caused flooding, debris 14 flows, and wet snow avalanches in the upper elevations of the Sierra Nevada. The origin of this 15 atmospheric river was super typhoon Jelawat, whose moisture remnants were entrained and 16 maintained by an extratropical cyclone in the northeast Pacific. This event was notable due to its 17 April occurrence, as six other typhoon remnants that caused heavy precipitation with high snow 18 levels (mean = 2.92 km) in the northern Sierra Nevada all occurred during October. 19


Introduction
The maritime midlatitude Sierra Nevada mountains provide key water resources, ecosystem services, and recreation opportunities for California and Nevada [1][2].The range receives over 70% of its annual precipitation during the cool season (October-April) with approximately 40-50% derived from atmospheric rivers [1,3].Atmospheric rivers (ARs) are narrow, elongated filaments of strong horizontal water vapor transport in the warm sector of an extratropical cyclone ahead of the cold front [4].ARs are fundamental components of western United States hydrometeorology and hydroclimate [1,3].The impacts of ARs can be positive [1,5] or negative [1,[5][6][7].
The orientation of the northwest-southeast trending Sierra Nevada (Figure 1) perpendicular to the west-southwest flow during storms leads to orographic precipitation enhancement when ARs interact with the topography [9][10].A meridional elevation gradient spans the range; the northern region is composed of lower characteristic elevations (median = 1.6 km above sea level) than the central (2.1 km) and southern (1.9 km) regions.This makes the northern region most susceptible to rain-on-snow flooding during storms with elevated snow levels and extreme precipitation [11][12].
However, when snow levels are extremely high (>2.5 km), large catchment areas of watersheds throughout the range will receive rainfall and contribute runoff from both direct precipitation and heat transfer-driven snowmelt [11].Because this precipitation is not being stored as snowpack, these warm storms create challenges for water and emergency managers who must handle the storm as a hazard rather than a resource [1,[13][14].It describes the extremely high snow levels associated with this storm and examines six other storms with likely typhoon moisture-remnant origins to further characterize snow levels during these events, one of which was previously studied [15].Last, some hydrogeomorphological impacts of the 5-7 April event are documented to provide examples of how typhoon-derived moisture influences natural hazards in mountain environments and may be recorded in the sedimentary record.

Materials and Methods
We focus on the period spanning 25 March -7 April 2018., or the period between when Jelawat was declared a super typhoon by the Japanese Meteorological Agency [16] and the most significant precipitation and impacts in the Sierra Nevada occurred.To identify and evaluate the possible linkage between moisture remnants of typhoon Jelawat and the landfalling AR, we use an experimental global product of total precipitable water (MIMIC-TPW2; [17]) that is based on morphological compositing of operational microwave-frequency satellite-based sensors [18].We also use 2.5° horizontal resolution NOAA non-interpolated outgoing longwave radiation (OLR) observations based upon Advanced Very High Resolution Radiometer data [19].OLR anomalies are reported as daily observations differenced from the 1979-1995 long term means averaged over a 22day period that brackets the period of interest by four days (i.e., 21 March -11 April).
Observations of precipitable water were acquired from ground-based Global Positioning Satellite data acquired near Half Moon Bay along the California coast (station OXMT; Figure 1) from 3 of 12 the COSMIC Suominet database [20].Hourly brightband elevations, or the elevation where snow turns to rain (typically 500 m below the 0°C isotherm), were derived from five snow level radars located west of the Sierra Nevada crest (Figure 1) [21][22][23][24].Climatological cool season (October-April) snow levels are provided for the Chico radar between 2002-2018.Six other fall-season typhoon remnant-derived precipitation events are included to evaluate their snow level characteristics in addition to those of Jelawat.
Hourly data were acquired from the California Data Exchange Center [25] and include observations of relatively unimpaired streamflow from the Merrimac gage on the Middle Fork of the Feather River (MFF) and alter-shielded tipping bucket precipitation at Bucks Lake (BKL) and Mammoth Pass (MHP; Figure 1).Daily streamflow observations were acquired from the United States Geological Survey [26] for the unimpaired gage on the Merced River at Pohono Bridge (MRP; Field observations of examples of hydrogeomorphological impacts of the event were recorded in high elevation regions of the Sierra Nevada between 15 April-10 May 2018 by Susan Burak, the Eastern Sierra Avalanche Center [27], Rob and Laura Pelewski [28] and the author.Observations of debris flows, high elevation flooding, and wet snow avalanches were conducted using ground-based and aerial photography.

Super Typhoon Jelawat
On 25 March 2018, super typhoon Jelawat can be identified as a region of negative OLR anomalies centered at 140°E, 8°N (Figure 2a) and as a region of precipitable water exceeding 65 mm and demonstrating radial inflow (Figure 3a).Between 25-29 March, Jelawat migrated northeastward with convective outflow to the northeast as indicated by increasing negative OLR anomalies (Figure 2b-c).This outflow merged with enhanced convection centered at 140°E, 8°N on 31 March (Figure 2d).As Jelawat dissipated, enhanced convection continued eastward across the Pacific along a band centered between 20°N and 30°N (Figures 2e-f).Negative OLR anomalies are observed on the equatorward side of an Aleutian low centered at the dateline (180°W; Figures 3c-d).Moisture from the remnants of Jelawat after 31 March and convective outflow from western tropical Pacific convection (Figures 2d-e) is evident in the precipitable water observations (Figures 3c-e).This moisture was entrained in the primary Aleutian low centered at the dateline (Figure 3d) before being zonally elongated by subsidiary lows moving eastward from the parent low (Figures 3e-f) towards North America.A canonical AR formed and made landfall between 6-7 April (Figures 3g-h

The 5-7 April Atmospheric River
Precipitable water observations at the California coast indicate that atmospheric moisture began increasing at 18Z on 5 April from less than 20 mm to in exceedence of 40 mm on 02Z 6 April (Figure 4a).Precipitable water values remained nearly continuously above 40 mm until approximately 13Z 7 April before rapidly declining.The peak observed precipitable water value of 51.9 mm was observed at 02Z 7 April (Figures 1b and 4a).Precipitation began in the northern Sierra Nevada at 04Z 6 April (Figure 4b).Two periods of intensified precipitation (>5 mm hr -1 ) are observed, between 17Z-23Z 6 April and between 08Z-13Z 7 April.Total event precipitation was 145 mm at Bucks Lake and 83 mm at Mammoth Pass.
During the onset of the event, snow levels ranged between 2.5 km and 3.2 km (lower in the north; Figure 4c).Coincident with the increase in precipitation rates after 17Z 6 April, an abrupt snow level rise of 1-1.4 km occurred at all stations except New Exchequer, which underwent a rise approximately five hours earlier.Snow levels remained above 3 km for the remainder of the event.hours.Contemporaneous with the first intense precipitation period (Figure 4b) and abrupt rise in snow levels (Figure 4c), flow rates accelerated between 18Z 6 April and 02Z 7 April.A second period with increasing flow rates occurred at 10Z 7 April with a peak flow rate of 375 cms at 14Z 7 April.This followed the second, more intense burst of precipitation (Figure 4b).

Climatological Context of Typhoon-Remnant Snow Levels
The distribution of cool season snow level observations (n = 5009 hours; Figure 5) at Chico demonstrates the extremely warm nature of this event.All 31 hours of observations from the 5-7 April event exceeded 2.75 km (Figure 5).This event included the greatest observed value at Chico (3.76 km) and 24% (12 hours) of snow levels exceeding the 99 th percentile (3.24 km).The snow levels during all seven typhoon-remnant events (n = 73 hours; Figure 5) have a mean of 2.92 km, which is approximately 1.2 km higher than the cool season mean (1.75 km).A two-tailed t-test indicates significant differences (p<0.001) between these means.90% of snow levels during these events exceeded the 85 th percentile (2.44 km) of the cool season distribution.These events compose 9% of the 85 th percentile snow levels while accounting for 1.3% of the total observations.The contribution from typhoon-remnant storms increases to 16% for hours exceeding the 95 th percentile and 27% for hours exceeding the 99 th percentile.

Hydrogeomorphological Impacts
Several high-elevation hydrogeomorphological impacts were documented following the event (see Figure 1 for location map).Anomalous high elevation (>2,800 m) streamflow near Sonora Pass is demonstrated by 1 m incut channels in avalanche and debris flow-covered creeks and the deposition of silts and sands on the snow surface as floodwaters retreated (Figure 6a).Vegetation debris and fine grained sediments were deposited onto the 1 m high cutbanks above the creek.
Several avalanches occurred, including a wet slab avalanche that occurred on a 3,000 m elevation, southwest aspect, low angle (10-15°) slope near Mammoth, California (Figure 6b).This avalanche triggered sympathetic failures on adjacent slopes and mobilized into a debris flow.The avalanche ran approximately 1,000 m, was approximately 180 m wide, and had an estimated mass between

Discussion
The evaluation of OLR and precipitable water over the tropical and extratropical North Pacific Ocean provides evidence for the linkage of super typhoon Jelawat to the AR that made landfall nearly two weeks later.The slowly eastward-propagating, deep northeast Pacific cyclone centered near the dateline (180 °W; visible in Figure 3c) and enhanced subtropical convection (Figure 2f) following the dissipation of Jelawat on 2 April in additional to quasigeostrophic forcing and a zonally-extended North Pacific jet stream facilitated eastward and poleward transport of the typhoon's remnant moisture, consistent with a case study of two October 2010 events [15].The zonal elongation noted in [15] is evident in the 5-7 April event (Figures 3g-h) and the six others studied and is supported by composites of total water vapor (not shown) from the European Center for Medium Range Forecasting ERA-Interim product [30].Precipitable water convergence along AR corridors and the role of strong frontogenesis along the poleward side of the AR [15], evidenced by low pressure circulations (Figures 3g-h and ERA-Interim products (not shown)) supports the notion of [31] that ARs are the "footprints" left behind from water vapor export from extratropical cyclones that converges along the cold front.However, the high snow levels associated with these events suggests that the air masses that accompany ARs will be characterized their environment of tropical origin.
This precipitation event demonstrated extremely high snow levels (Figure 4), which are supported by comparison to the distribution of cool season snow levels (Figure 5).These high snow levels are consistent with other ARs that entrained typhoon moisture remnants, including events in October 2008 (Typhoon Jangmi), 2009 (Typhoon Melor), 2010 (Typhoons Megi and Chaba; [15]), and 2016 (Typhoons Songda and Haima).The finding that ARs with typhoon-remnant origins demonstrate extremely high snow levels (Figure 5) is consistent with the positive linear relationship between precipitable water and snow level identified during the winter season by [12].Because this event occurred in April and a ripened, spring season antecedent snowpack was present, the impacts of the high snow levels and heavy precipitation were amplified compared to the October events.
Heavy rainfall and melting snow contributed to the extreme runoff and streamflow and the magnitude of observed mass movements (Figure 6).The observations from Pine Creek, California of wet snow avalanches mobilizing debris flows that were ultimately issued upon ice-covered lakes recorded provides evidence supporting the idea that extreme hydrometeorological events can be preserved in the sedimentary record of high alpine lakes [32][33][34][35].
Typhoon-remnants entrained in ARs appear to be more common during October.October is the start of the western United States water year, which spans 1 October-30 September.The April event provides a counter-example to this perceived seasonality.Heavy October precipitation events create positive water year precipitation anomalies, which can facilitate the onset of warm snow drought conditions if rain occurs rather than snow.Warm snow droughts are characterized by near or above normal precipitation but below normal snowpack [36] and can imply that water is not being stored in the snowpack and must be managed through legally-constrained reservoir operations [14].Other early water year warm snow droughts related to typhoons include the Columbus Day Storm in October 1962, which had origins as Typhoon Freda, and the water year 2017 warm snow drought initiated by Typhoons Songda and Haima [36].
Early water year typhoon-related precipitation events can also be beneficial.In the higher elevations of the Sierra Nevada (i.e., above 3 km) these storms can create substantial early season snowpack.At lower elevations, they can provide respite from the extended summer drought conditions that characterize California's Mediterranean climate [37].Heavy precipitation can cause streamflows to rise above baseflow, which benefits ecosystems.The wetting of soils and vegetation can reduce wildfire hazard when fuel moistures are typically at their climatologically lowest values [38].Conversely, the abundant moisture associated with these storms (Figure 4a) can generate shortduration, intense precipitation (Figure 4b), which can lead to damaging post-fire debris flows [7].This created an environment conducive to high elevation flooding and the formation of debris flows and wet snow avalanches.The deposition of debris upon frozen lakes provides direct evidence supporting [32] that lacustrine sediments in Sierra Nevada alpine environments can record extreme hydrometeorological events.Typhoon-derived precipitation appears to be more common in October in the Sierra Nevada, yet the 5-7 April event demonstrated different impacts due to the presence of an antecedent snowpack.Typhoon-remnant storms tend to be warm, with mean snow levels that are equivalent to the 85 th percentile of cool season snow levels.
Continuing work seeks to develop a robust climatology of the contributions of typhoon-remnant moisture to early water year precipitation.Under projected global and regional climate change, California's wet season is expected to become narrower and more extreme [39], which will strain existing infrastructure and water management institutions [14,40].Warming temperatures and shifting precipitation regimes will drive increases in the fire season length [41] and intensify drought severity [37,39].While projections vary on changes in Pacific tropical cyclogenesis frequency [42][43], tropical cyclones are expected to undergo a poleward shift [43] that may favor increases in eventual North American landfall.Future contributions of remnant typhoon moisture to early water year precipitation could become more important in relieving drought conditions and mitigating wildfire hazard along the west coast of North America, both of which have positive impacts for ecosystems and human life and property.

Funding:
The project described in this publication was supported by Grant Number G14AP00076 from the United States Geological Survey.Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the USGS.This manuscript is submitted for publication with the understanding that the United States Government is authorized to reproduce and distribute reprints for governmental purposes.

Figure 1 )
Figure 1) between 1 October 1916 -30 April 2018 to provide long-term climatological context of the runoff during the 5-7 April event.

Preprints
(www.preprints.org)| NOT PEER-REVIEWED | Posted: 24 May 2018 doi:10.20944/preprints201805.0333.v1Peer-reviewed version available at Atmosphere 2018, 9, 233; doi:10.3390/atmos90602336 of 12The highest values (>3.75 km) were observed in the southern portion of the Sierra Nevada (i.e., New Exchequer and Pine Flat Dam; Figure4c) between 18Z 6 April and 18Z 7 April.Shasta Dam observed an approximate 1 km fall in snow levels after 18Z 7 April, and although Chico and Colfax declined approximately 0.3 km during this time, snow levels rebounded to above 3.6 km during the second wave of intensified precipitation (6-10 mm hr -1 ; Figure4b).The greatest observed snow level rangewide was observed at New Exchequer (4.0 km); this station had 12 hours of snow level observations exceeding 3.7 km.

Figure 4 .
Figure 4. (a) Time series of precipitable water (mm) near Half Moon Bay; (b) observed streamflow (cms) on the Middle Fork of the Feather River; (c) brightband-derived snow levels at five snow level radars upstream of the Sierra Nevada crest (km; ordered from north to south) and (d) precipitation observations (mm) at Bucks Lake, California.

Figure 5 .
Figure 5. Histogram of hours of observed snow levels (km) at Chico, California for all cool seasons (October-April; blue bars), seven typhoon events (red bars), and the 5-7 April 2018 event (black bars).

10 3 -
10 4 t.Other wet slab avalanches that subsequently mobilized debris flows were observed to issue debris onto the surfaces of high elevation frozen lakes, including Lower and Upper Pine Lakes (Figures6c-d) in the Pine Creek drainage near Bishop, California.Flooding of high elevation meadows and the Tuolumne River in Tuolumne Meadows was observed immediately following Preprints (www.preprints.org)| NOT PEER-REVIEWED | Posted: 24 May 2018 doi:10.20944/preprints201805.0333.v1Peer-reviewed version available at Atmosphere 2018, 9, 233; doi:10.3390/atmos9060233 the event (Figures 6e-f).Mining access roads in Pine Creek Canyon were damaged by debris chutes forming from wet snow avalanches and debris flows (Figure 6g).The high elevation hydrological response to this event was well-captured by the Merced River.The streamflow observation of 257 cms on 7 April was the highest April flow on record and the 10 th largest flow since observations began in 1916 (37,102 daily observations).The severity of impacts due to Merced River flooding closed Yosemite National Park for two days [29].

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 24 May 2018 doi:10.20944/preprints201805.0333.v1
This communication provides examples of how entrainment of typhoon-originating moisture into an atmospheric river (AR) can impact hydrologic systems and natural hazards in mountain areas far removed from the region of tropical cyclogenesis.An analysis of outgoing longwave radiation and precipitable water provided evidence that super typhoon Jelawat contributed moisture to an AR that impacted the Sierra Nevada between 5-7 April 2018.This event included heavy precipitation (>80 mm) with multiple hours of extremely high (>3.75 km) snow levels throughout the Sierra Nevada.