The snow-shadow effect of Sado Island on Niigata City and the coastal plain

概要

Received: 26 February 2023

Revised: 3 July 2023

Accepted: 6 July 2023

DOI: 10.1002/asl.1182

RESEARCH ARTICLE

The snow-shadow effect of Sado Island on Niigata City
and the coastal plain
Hiroyuki Kusaka 1

|

Nobuyasu Suzuki 2 |

Masato Yabe 3 | Hiroki Kobayashi 3

1

Center for Computational Sciences,
University of Tsukuba, Tsukuba, Japan

2

Graduate School of Science and
Technology, University of Tsukuba,
Tsukuba, Japan

3

Graduate School of Life and
Environmental Sciences, University of
Tsukuba, Tsukuba, Japan
Correspondence
Hiroyuki Kusaka, Center for
Computational Sciences, University of
Tsukuba, Tsukuba, Japan.
Email: kusaka@ccs.tsukuba.ac.jp

Funding information
Environmental Restoration and
Conservation Agency; Environment
Research and Technology Development
Fund, Grant/Award Number:
JPMEERF20232003

Abstract
Japan's Hokuriku region, near the Sea of Japan, typically experiences heavy
snowfall; however, Niigata City, the largest city on the Sea of Japan side, experiences lower levels of snowfall than neighbouring cities. This study investigates the snow-shadow effect of Sado Island on snowfall in Niigata City,
located 45 km away leeward. Statistical analysis of long-term radar data for
10 winters showed that snow-shadow effects in the Niigata plain occurred in
151 (80%) of the 188 cases, during which a strong approaching wind reached
the island. The location of this snow-shadow effect was always downwind of
Sado Island and depended on the wind direction. Numerical experiments using
the Weather Research and Forecasting model predicted that snowfall over the
Niigata Plain would be lighter with the island than without it. Additionally,
the snow-shadow effect occurs in areas more than 150 km downwind. The
experiments showed that Sado Island reduces heat fluxes from the sea surface
by weakening leeward winds. At the same time, the horizontal wind convergence downwind is weakened. Meanwhile, the orographic snowfall over Sado
Island reduces the amount of water vapour, cloud water and cloud ice over the
leeward sea. Therefore, Sado Island prevents cloud lines from redeveloping
over the leeward sea and can further reduce snowfall over the leeward plain,
including in Niigata City.
KEYWORDS

Hokuriku region, orographic precipitation, Sado Island, sea-effect snowfall, snow shadow,
winter monsoon

1 | INTRODUCTION
The Hokuriku region, situated on the Sea of Japan side
of the country, often experiences heavy snowfall
despite its location at approximately 38 N latitude
(e.g., Steenburgh & Nakai, 2020; Yoshino, 1977). This
is due to the air mass transformation effect caused by
the cold north-westerly monsoon passing over the

warmer Sea of Japan, leading to the development of
cumulonimbus clouds and resulting in snowfall
(e.g., Kato & Asai, 1983; Manabe, 1957; Nakamura &
Asai, 1985; Ninomiya, 1964). The snowfall mechanism
in this region is similar to that observed along the
Great Lakes coast of North America (e.g., Andersson &
Gustafsson, 1994; Niziol et al., 1995; Passarelli Jr &
Braham Jr, 1981).

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided
the original work is properly cited.
© 2023 The Authors. Atmospheric Science Letters published by John Wiley & Sons Ltd on behalf of Royal Meteorological Society.
Atmos Sci Lett. 2023;e1182.
https://doi.org/10.1002/asl.1182

wileyonlinelibrary.com/journal/asl

1 of 12

KUSAKA ET AL.

F I G U R E 1 (a) Domains of numerical simulations. Overview of the four nested domains. (b) Magnified view of Domain 4 (D04).
Ak, Sa, Ni, Ta and To indicate Akita, Sakata, Niigata, Takada and Toyama local meteorological observatories of Japan Meteorological
Agency (JMA), respectively. Wa indicates the JMA's Wajima Aerological Observatory. The values indicate the climatological normal of
snowfall amount in winter (December to February). The climatology is defined by 1991–2020.

Snowfall in the Hokuriku region is affected by the
Japan-Sea Polar-airmass Convergence Zone, the strength of
the monsoon and the conditions of the cold air above
(e.g., Nagata, 1991; Nagata et al., 1986; Shimizu et al., 2017;
Shinoda et al., 2021; Yoshizaki et al., 2004). Snowfall primarily occurs in the mountains during strong monsoons and in
the plains when a cold vortex is present over the Sea of Japan
(e.g., Akiyama, 1981a, 1981b; Eito et al., 2005; Iwamoto
et al., 2008; Nakai et al., 2005; Ohigashi & Tsuboki, 2005).
Although the Hokuriku region typically experiences
heavy snowfall, Niigata City, located on the Niigata Plain
and the largest city on the Sea of Japan side, experiences
less snowfall than other cities (Figure 1; e.g., Matsumoto, 1967; Veals et al., 2019). One possible explanation for
Niigata City experiencing less snowfall might be due to its
nature being the largest plain on the Sea of Japan side and
far from the mountains. Additionally, the city is outside the
area where the Japan-Sea Polar-airmass Convergence Zone
reaches a high frequency (e.g., Shimizu et al., 2017; Shinoda
et al., 2021). It remains unclear whether additional factors
affect snowfall levels in this area.
Figure 2 shows a weather chart, winds at the 850 hPa
level, and precipitation distribution on a typical day with
little snow over most of the Niigata Plain. Matsumoto
(1967) and Yagi and Uchiyama (1983) investigated cloud
movement and suggested that the reduced snowfall pattern leeward of Sado Island was formed by north-westerly
winds bypassing the island. However, Veals et al. (2019)
found that the snowfall amount downwind of Sado Island
is less than its surroundings regardless of approaching
wind direction and speed. The results of the days with

strong winds, when air flows over the island, suggest that
the factors proposed by Matsumoto (1967) and Yagi and
Uchiyama (1983) may not be the primary reasons. It is
still unclear how Sado Island reduces snowfall in cities
45 km away from the island.
Sado Island is situated approximately 45 km northwest of Niigata City and has an area of approximately
855 km2 (Figures 1 and S1). The summit of the largest
mountain on the island is 1172 m above sea level. When
the northwest monsoon is strong (e.g., wind speed
>9 m/s and dividing streamline height is lower than the
ground level), the approaching air is expected to flow
over Sado Island. If wind speeds are very strong (or, more
strictly, if the mountain Froude number is very large),
snow can flow over the mountain range and reach its
downwind side as spill-over (e.g., Sinclair et al., 1997). In
contrast, winds below the dividing streamline height can
travel around mountains and reconverge downwind,
where clouds form again. However, due to the relatively
short distance between Sado Island and the Niigata Plain,
which is only 45 km, it is uncertain whether clouds can
redevelop and produce snowfall in the Niigata Plain. It is
also unclear if Sado Island has a snow-shadow effect that
could reduce snowfall in the region.
Climatology research has demonstrated rain-shadow
effects in large mountain ranges, such as the Alps, Olympic, Sierra Nevada, and Andes Mountains (e.g., Ahrens,
2001; Frei & Schär, 1998; Kaplan et al., 2009; Medina &
Houze Jr, 2003; Siler & Durran, 2016; Underwood
et al., 2009; Whiteman, 2000). Theoretical studies of orographic precipitation were conducted in the 2000s

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

2 of 12

3 of 12

F I G U R E 2 (a) Surface weather chart at 12 UTC on 26 December 2008. (b) Geopotential height (contours at 30-m intervals) and
horizontal wind (wind barbs) at 850 hPa level. (c) 12-h accumulated precipitation from 00 UTC to 12 UTC with ground-based radar. The
thick line indicates the coastline and contours indicate elevation at intervals of 500 m. The full and half barbs represent 10 and 5 kt/h,
respectively.

(e.g., Smith & Barstad, 2004), and more recently, studies
have been conducted using long-term data and numerical
models. Siler and Durran (2016) identified differences in
the dynamic mechanisms of strong and weak rainshadows within the Cascade Range.
Rain- or snow-shadow effects are typically observed in
large mountain ranges on continents and the mountains on
relatively large islands. During the northwest winter monsoon in Japan, the windward side of the backbone mountain
range (Sea of Japan side) is often heavily snow-covered,
whereas the leeward side (Pacific Ocean side) typically has
clear skies. This is a typical snow-shadow effect. Rainshadow effects have been observed in Wales and the Peak
Districts of the United Kingdom (e.g., Sawyer, 1956;
Stockham et al., 2018), Taiwan (e.g., Yeh & Chen, 1998),

Luzon Island, Philippines (e.g., Akasaka et al., 2007; Chang
et al., 2005; Pullen et al., 2015), the Southern Alps of New
Zealand (e.g., Chater & Sturman, 1998; Sinclair et al., 1997)
and Sri Lanka (e.g., Puvaneswaran & Smithson, 1991).
Rain-shadow effects are typically observed downwind of
islands. In Hawaii, which has mountains reaching heights
of up to 4 km, precipitation amount and frequency decrease
by approximately 40 km downwind of the island when
trade winds are dominant (e.g., Kidd & McGregor, 2007;
Nullet & McGranaghan, 1988). Although rain-shadow
effects have been studied in various locations worldwide, it
has not been extensively examined on small mountains
without high elevations, such as Sado Island in Japan.
It remains unclear whether a small mountain on a
small island can block strong winds and reduce snowfall

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

KUSAKA ET AL.

KUSAKA ET AL.

on a leeward plain located 45 km away. To address this
knowledge gap, we investigated the snow-shadow effect of
Sado Island on snowfall in Niigata City. Our study aims to
improve the understanding of mountain meteorology in
terms of the snow-shadow effect on small islands rather
than the large rain-shadow effect on medium-to-largesized mountains investigated previously.

2 | DATA AND METHODS

images created from geostationary satellites and radar
data. Specifically, a qualified event met the above criteria
for at least one of the 3-hourly time bins: 0–2, 3–5, 6–8,
9–11, 12–14, 15–17, 18–20 and 21–23 UTC. In the Sea of
Japan, the wind direction and cloud bands may be parallel
(L-mode) or orthogonal (T-mode), and their selection
conditions are complex and involve vertical wind shear,
atmospheric stability and buoyancy (e.g., Asai, 1972; Eito
et al., 2010). Therefore, we visually confirmed the events
and did not use an automated approach, using quantitative criteria for atmospheric conditions.

2.1 | Data used in this study
Ground-based radar data from the Japan Meteorological
Agency (JMA) were used to understand the spatial patterns
of snowfall in the study area. The observational sampling
interval and spatial resolution were 10 min and 1 km,
respectively. Cloud image data measured by the Himawari
satellite, operated by the JMA, were also used. The 12-hourly
sonde data observed at the JMA Wajima Aerological Observatory and the 6-hourly surface weather charts generated
by the JMA were used to understand the synoptic weather
conditions and prevailing winds around Niigata City.
The 6-hourly National Center for Environmental Prediction Final (NCEP-FNL) data and daily NCEP-real-time
global sea surface temperature data were used to create
the initial and boundary conditions for the numerical
simulations using the Weather Research and Forecasting
(WRF) model.

2.2.2 | Configurations of numerical
experiments
The WRF model version 4.2.2 (Skamarock et al., 2019) was
used for numerical simulations. The four nesting domains
of the WRF model are shown in Figure 1a. The horizontal
grid spacings for the domains D01, D02, D03 and D04 were
27, 9, 3 and 1 km, respectively. There were 61 vertical
layers. The other configurations are listed in Table 1.
Numerical simulations were conducted for 12 typical
precipitation band events listed in Table 2. A numerical
experiment with real terrain is referred to as CTRL.
Sensitivity experiments were conducted to evaluate the
effect of Sado Island on the precipitation around Niigata
City (Table 3). In Case No_SD, simulations were performed
TABLE 1

Model configuration.

2.2 | Methods
Model

2.2.1

| Statistical analysis

We investigated how snowfall distribution around Niigata
City is affected by the island. As dates and times for the
analysis, we selected snowfall events that occurred during
10 winters (December–February 2005–2014) and met the
following three conditions: (i) the typical winter-type pressure pattern of Japan, with an anticyclone to the west of
Japan and a cyclone to the east. (ii) A wind direction at
the 850-hPa level observed using the GPS-Sonde from the
Wajima Aerological Observatory between west and northwest, with a wind speed of 9 m/s or higher; where 9 m/s
is the average wind speed when the height of the dividing
streamline is at ground level. Because wind data at the
850-hPa level were available only every 12 h, data at times
other than the observation time were assumed to be the
same as those at the nearest observation time. (iii) Cloud
lines and precipitation bands were present on the windward side of Sado Island in the same direction as the wind
at 850 hPa for >3 h. This was visually examined using

Advanced research WRF
version 4.2.2

Horizontal grid-spacing

27 km (D01), 9 km (D02), 3 km
(D03), 1 km (D04)

Number of vertical layers

61

Initial and boundary
conditions

NCEP-FNL and RTG-SST data

Topography

Digital national land
information data by the
geospatial information
authority of Japan

Land use

USGS data

Short-wave radiation

Simple radiation
(Dudhia, 1989)

Long-wave radiation

RRTM (Mlawer et al., 1997)

Land surface

Noah-LSM (Chen &
Dudhia, 2001)

Boundary layer turbulence

YSU (Hong et al., 2006)

Cloud microphysics

WSM6 (Hong et al., 2004)

Cumulus parameterization

Kain-Fritch (Kain, 2004) only
for D01 and D02

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

4 of 12

5 of 12

T A B L E 2 List of 12 typical events used for the numerical
experiments.
Events
From 12 UTC 27th to 00 UTC on 28 December 2005
From 15 UTC 5th to 00 UTC on 6 January 2006
From 12 UTC 8th to 00 UTC on 9 January 2007
From 12 UTC 15th to 00 UTC on 16 February 2008
From 00 UTC to 12 UTC on 26 December 2008
From 03 UTC to 12 UTC on 21 December 2009
From 18 UTC 25th to 00 UTC on 26 December 2011
From 03 UTC to 09 UTC on 29 January 2012
From 15 UTC 18th to 00 UTC on 19 December 2012
From 03 UTC to 12 UTC on 4 January 2013
From 18 UTC, 16th to 00 UTC on 17 February 2013
From 03 UTC to 12 UTC on 17 January 2015

T A B L E 3 Configurations of topography in the numerical
experiments.
Name of
experimental
cases

Topography

CTRL

Real topography

No_SD

Topography without Sado Island

NE_SD

Sado Island was moved to 150 km northeast
of its actual location while maintaining the
distance between the Island and the
mainland

NW_SD

Sado Island was moved to 200 km upwind
(northwest) from its actual position

by removing Sado Island. In Case NE_SD, Sado Island was
moved 150 km northeast of its actual location while maintaining the distance between the island and the mainland.
In this case, Sado Island is located upwind of the plain,
where the Sakata local meteorological observatory is situated. In Case NW_SD, Sado Island was shifted 200 km
upwind (northwest) from its actual position, away from
Niigata City. These numerical experiments employed
the same settings as the CTRL experiment, with the only
difference being the location of Sado Island.

3 | R E SUL T S
3.1 | Statistical and composite analyses
A total of 564 h of precipitation bands met all three
conditions described in Section 2.1. In 453 (i.e., 151  3-h

F I G U R E 3 Composites of observed 3-h accumulated
precipitation amounts for (a) westerly winds (45 h), (b) westnorth-westerly winds (258 h) and (c) north-westerly winds events
(150 h). Grey areas indicate areas out of radar observation.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

KUSAKA ET AL.

samples) (80%) of these 564 h (i.e., 188  3-h samples),
the precipitation in Niigata City was lower than that in
the surrounding areas. Figure 3 shows a precipitation

KUSAKA ET AL.

composite for each event, classified according to wind
direction. In all figures, precipitation was low downwind
of Sado Island, regardless of the wind direction.

F I G U R E 4 (a) Observed 3-h accumulated precipitation. (b) Simulated 3-h accumulated precipitation for Case CTRL. (c) Simulated 3-h
accumulated precipitation for Case No_SD. (d) Differences in 3-h accumulated precipitation between Cases CTRL and No_SD. Only regions
with a 95% confidence level in the Welch's t-test are shown in Figure 4d. The values for all figures were averaged for the 12 typical events
listed in Table 2.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

6 of 12

Specifically, the snow-shadow effect was observed in different areas for westerly winds compared to other wind
directions, such as north-westerly and westnorth-westerly winds. These results suggest that a snowshadow effect occurs on Sado Island.

7 of 12

3.2 | Numerical experiments
Figure 4a,b depict the spatial distributions of 3-h precipitation from observations and numerical experiments
(Case CTRL), respectively. All values in Figure 4 are

F I G U R E 5 The 3 h accumulated precipitation for (a) Case NE_SD and (c) Case NW_SD. (b) Differences between Cases NE_SD and
No_SD. (d) Differences between Cases NW_SD and No_SD. Only regions with a 95% confidence level in the Welch's t-test are shown in
Figure 5b,d. The values for all figures were averaged for the 12 typical events listed in Table 2.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

KUSAKA ET AL.

averaged for the simulated results of the 12 typical events
listed in Table 2. In the CTRL experiment, the WRF
model reproduced a critical feature of the observations:
the appearance of a minimum snowfall downwind of
Sado Island. Interestingly, both the observations and simulations showed that the precipitation minima extended
not only over the Niigata Plain where Niigata City is
located but also over the mountain slopes behind the plain;
in contrast, in the No_SD experiment (Figure 4c), there
was no substantial difference between the snowfall in
Niigata City and the surrounding areas in the plain, and
the snowfall in the Niigata Plain was comparable to that in
other plains. The impact of Sado Island on the precipitation
patterns is shown in Figure 4d. As depicted in this figure,

KUSAKA ET AL.

large precipitation anomalies were found only downwind
of Sado Island. Although the model bias is found in the
mountain areas, the results presented in Figure 4d can be
considered acceptable because of the following reasons:
precipitation at high elevations estimated from the radar is
likely underestimated (e.g., Veals et al., 2019), and model
bias can be cancelled out and reduced when comparing the
experiments CTRL and No_SD.
To investigate the hypothesis that Niigata City experiences low snowfall levels because of its location on a relatively large plain near the Sea of Japan, a simulation was
conducted in which Sado Island was moved upwind of a
smaller plain, specifically to the northeast (Case NE_SD).
The results are shown in Figure 5a. A minimal snowfall

F I G U R E 6 Distribution of wind speed at 10 m, heat fluxes from the surfaces and horizontal convergence of winds at 950 hPa. (a) Wind
speed in Case CTRL, (b) wind speed in Case No_SD and (c) differences in wind speed between Cases CTRL and No_SD. (d) Surface heat
fluxes in Case CTRL, (e) surface heat fluxes in Case No_SD and (f) differences in surface heat fluxes between Cases CTRL and No_SD.
(g) Horizontal convergence of winds in Case CTRL, (h) horizontal convergence of winds in Case No_SD and (i) differences in horizontal
convergence of winds between Cases CTRL and No_SD. The values for all figures were averaged for the 12 typical events listed in Table 2.
Vector arrows in (a) and (b) indicates wind direction only.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

8 of 12

area appeared downwind of the new Sado Island but not
around Niigata City. As shown in Figure 5b, there was a
substantial decrease in snowfall in the plain downwind
of the new Sado Island compared to that seen for Case

9 of 12

No_SD. These results show that snowfall decreases under
the influence of Sado Island, regardless of whether the
downwind area of Sado Island is a large or small plain.
Furthermore, Figure 5b shows that Sado Island

F I G U R E 7 Vertical cross-section along A–B line (shown in Figure S4) of winds, relative humidity and mixing ratio of precipitating and
in cloud liquid and frozen particles over 0.001 g kg 1. Here, precipitating and in-cloud liquid and frozen particles are defined by the sum of
cloud water, cloud ice, rain, snow and graupel in the microphysics scheme. (a) Winds and relative humidity in Case CTRL, (b) winds and
relative humidity in Case No_SD and (c) differences in winds and relative humidity between Cases CTRL and No_SD. (d) Mixing ratio of
precipitating and in cloud liquid and frozen particles (over 0.001 g kg 1) in Case CTRL, (e) mixing ratio of precipitating and in cloud liquid
and frozen particles (over 0.001 g kg 1) in Case No_SD and (f) difference in the mixing ratios between Cases CTRL and No_SD. The values
for all figures were averaged for the 12 typical events listed in Table 2. Solid lines in (d) and (e) indicate the cloud water mixing ratio greater
than 10 3 g kg 1. The solid and dashed lines in (f) indicate the difference in cloud water mixing ratio between CTRL and No_SD greater
than 10 3 g kg 1 and less than 10 3 g kg 1, respectively.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

KUSAKA ET AL.

influences not only the plains but also the mountainous
areas downwind.
To understand how far the influence of Sado Island
extends, we moved Sado Island upwind, specifically to
the northwest 200 km (Case NW_SD). The results show
that a minimal snowfall area appeared downwind of
Sado Island (Figure 5c). The difference between cases
NW_SD and No_SD illustrated in Figure 5d indicates
that Sado Island can reduce snowfall by more than
150 km downwind; thus, the potential impact on Sado
Island is more than 125-fold farther than the elevation
of Sado's largest mountain. This is probably because the
change in winds around Sado Island caused a convergence line shift, which also shifted the snowfall area
over the plains.
The results of precipitation simulations are often
influenced by the selected physics schemes. Therefore,
we validated the CTRL and No_SD results using physics
scheme ensemble experiments. Ensemble experiments
were conducted with 12 ensemble members created using
four different cloud microphysics schemes, WSM6 (Hong
et al., 2004), WDM6 (Sunny Lim & Hong, 2010), Thompson (Thompson et al., 2008) and Morris (Morrison
et al., 2009) schemes and three different planetary boundary layer schemes, YSU (Hong et al., 2006), MYJ
(Janjic, 1994) and ACM2 (Pleim, 2007a, 2007b). The configurations in the ensemble experiments with and without
Sado Island were the same as those in the CTRL and
No_SD experiments described in Section 2.2, except for
the selected physical schemes. The experiments were conducted for the 26 December 2008 event. The ensembleaveraged results from the 12 members shown in Figure S2
are similar to those of the CTRL and No_SD experiments
shown in Figure 4.
The results of the precipitation simulations were influenced by both the initial/boundary conditions selected as
well as the physics schemes. Therefore, we conducted
CTRL and No_SD experiments using the initial/boundary
conditions created from the ERA5 data instead of the
NCEP-FNL data. The configurations and 12 target events
in the experiments were the same as those in the CTRL
and No_SD experiments described in Section 2.2. The
results from the experiments shown in Figure S3 are
almost identical to those in Figure 4. The results of
the physics ensemble and initial/boundary condition
ensemble experiments increased the robustness of the
conclusions drawn in Section 3.2.
Finally, to examine the factors contributing to the
snow-shadow effect, we examined how the winds and
fields of water vapour and cloud water and ice content
differ depending on the presence or absence of Sado
Island. Figure 6 indicates that Sado Island weakens the
winds over the leeward ocean, consequently reducing
the heat flux from the sea surface to the atmosphere. The

KUSAKA ET AL.

reduced heat fluxes were primarily caused by a reduction
in the latent heat flux (Figure omitted). Additionally,
Sado Island weakened horizontal wind convergence
downwind. Another important aspect is that Sado Island
produces orographic snowfall, reducing water vapour and
cloud water and ice content over the leeward ocean
(Figure 7). Thus, we believe that Sado Island prevents
precipitation band from redeveloping over the leeward
ocean and plains, reducing snowfall.

4 | CONCLUSIONS
In this study, we examined whether Sado Island reduced
precipitation on Niigata City, located 45 km downwind of
the island (snow-shadow effect), by analysing long-term
radar data and performing numerical experiments. The
results of the radar data analysis for the past 10 winters
showed that the snow-shadow effect occurred in
453 (151 3-h samples, 80%) of 564 h (188 3-h samples)
when the monsoon reached Sado Island. The location of
this snow-shadow effect depends on the wind direction
and was always observed downwind of Sado Island.
Numerical experiments using the WRF model indicated that snowfall over the Niigata Plain would be less
with Sado Island than without it. When the island was
moved 200 km windward, the snow-shadow effect
appeared more than 150 km downwind of the sea.
These results indicate that even small islands without
high mountains, such as Sado Island, can reduce snowfall
over leeward plains.
The numerical experiments also showed that (i) Sado
Island reduces heat fluxes from the sea surface by weakening leeward winds, (ii) horizontal wind convergence
is weakened in the leeward plain of Sado Island and
(iii) Sado Island produces orographic snowfall and
reduces the amount of water vapour and cloud water and
ice over the leeward sea. These factors prevent cloud lines
from redeveloping over the leeward ocean and create the
snow-shadow effect on Sado Island.
However, there are several other reasons for low
snowfall in Niigata City, as discussed in Section 1. Thus,
the percentage of the total contribution from the snowshadow effect on Sado Island remains unclear. To understand this, long-term climate downscaling experiments
for 10–20 years with and without islands are necessary.
AUTHOR CONTRIBUTIONS
Hiroyuki Kusaka: Conceptualization; formal analysis;
funding acquisition; methodology; project administration; resources; supervision; validation; writing – original
draft; writing – review and editing. Nobuyasu Suzuki:
Formal analysis; methodology; validation; visualization;
writing – original draft; writing – review and editing.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

10 of 12

11 of 12

Masato Yabe: Conceptualization; formal analysis;
methodology. Hiroki Kobayashi: Formal analysis;
methodology; writing – original draft.
ACK NO WLE DGE MEN TS
This study was supported by the Multidisciplinary
Cooperative Research Program (Oakforest-PACS) of
the Center for Computational Sciences, University of
Tsukuba. MTSAT-1R/2 gridded data were distributed by
the Centre for Environmental Remote Sensing (CEReS),
Chiba University, Japan.
DATA AVAILABILITY STATEMENT
The data are available from the corresponding author
upon request.
ORCID
Hiroyuki Kusaka
7179

https://orcid.org/0000-0002-0326-

R EF E RE N C E S
Ahrens, C.D. (2001) Essential of meteorology: an invitation to
atmosphere, 3rd edition. Pacific Grove, CA: Brooks/Core Publishing CO., p. 465.
Akasaka, I., Morishima, W. & Mikami, T. (2007) Seasonal march
and its spatial difference of rainfall in The Philippines. International Journal of Climatology, 27, 715–725.
Akiyama, T. (1981a) Time and spatial variations of heavy snowfalls
in the Japan Sea coastal region. Part I. Principal time and space
variations of precipitation described by EOF. Journal of the
Meteorological Society of Japan, 59, 578–590.
Akiyama, T. (1981b) Time and spatial variations of heavy snowfalls in
the Japan Sea coastal region. Part II. Large-scale situations for
typical spatial distributions of heavy snowfalls classified by EOF.
Journal of the Meteorological Society of Japan, 59, 591–601.
Andersson, T. & Gustafsson, N. (1994) Coast of departure and coast
of arrival: two important concepts for the formation and structure of convective snowbands over seas and lakes. Monthly
Weather Review, 122, 1036–1049.
Asai, T. (1972) Thermal instability of a shear flow turning the direction with height. Journal of the Meteorological Society of Japan,
50, 525–532.
Chang, C.P., Wang, Z., McBride, J. & Liu, C.H. (2005) Annual cycle
of Southeast Asia-Maritime Continent rainfall and the asymmetric monsoon transition. Journal of Climate, 18, 287–301.
Chater, A.M. & Sturman, A.P. (1998) Atmospheric conditions
influencing the spillover of rainfall to lee of the Southern
Alps, New Zealand. International Journal of Climatology, 18,
77–92.
Chen, F. & Dudhia, J. (2001) Coupling an advanced landsurface/hydrology model with the Penn State/NCAR MM5
modeling system. Part I: model description and implementation. Monthly Weather Review, 129, 569–585.
Dudhia, J. (1989) Numerical study of convection observed during
the winter monsoon experiment using a mesoscale twodimensional model. Journal of the Atmospheric Sciences, 46,
3077–3107.

Eito, H., Kato, T., Yoshizaki, M. & Adachi, A. (2005) Numerical
simulation of the quasi-stationary snowband observed over the
southern coastal area of the sea of Japan on 16 January 2001.
Journal of the Meteorological Society of Japan, 83, 551–576.
Eito, H., Murakami, M., Muroi, C., Kato, T., Hayashi, H.,
Kuroiwa, H. et al. (2010) The structure and formation mechanism of transversal cloud bands associated with the Japan-sea
polar-airmass convergence zone. Journal of the Meteorological
Society of Japan, 88, 625–648.
Frei, C. & Schär, C. (1998) A precipitation climatology of the Alps
from high-resolution rain-gauge observations. International
Journal of Climatology, 18, 873–900.
Hong, S.Y., Dudhia, J. & Chen, S.H. (2004) A revised approach to ice
microphysical processes for the bulk parameterization of clouds
and precipitation. Monthly Weather Review, 132, 103–120.
Hong, S.Y., Noh, Y. & Dudhia, J. (2006) A new vertical diffusion
package with an explicit treatment of entrainment processes.
Monthly Weather Review, 134, 2318–2341.
Iwamoto, K., Nakai, S. & Sato, A. (2008) Statistical analyses of
snowfall distribution in the Niigata area and its relationship to
the wind distribution. Scientific Online Letters on the Atmosphere: SOLA, 4, 45–48.
Janjic, Z.I. (1994) The step–mountain eta coordinate model: further
developments of the convection, viscous sublayer, and turbulence closure schemes. Monthly Weather Review, 122, 927–945.
Kain, J.S. (2004) The Kain-Fritsch convective parameterization: an
update. Journal of Applied Meteorology, 43, 170–181.
Kaplan, M.L., Adaniya, C.S., Marzette, P.J., King, K.C.,
Underwood, S.J. & Lewis, J.M. (2009) The role of upstream
midtropospheric circulations in the Sierra Nevada enabling leeside (spillover) precipitation. Part II: a secondary atmospheric
river accompanying a midlevel jet. Journal of Hydrometeorology, 10, 1327–1354.
Kato, K. & Asai, T. (1983) Seasonal variations of heat budgets in
both the atmosphere and the sea in the Japan Sea area. Journal
of the Meteorological Society of Japan, 61, 222–238.
Kidd, C. & McGregor, G. (2007) Observation and characterisation
of rainfall over Hawaii and surrounding region from the tropical rainfall measuring Mission. International Journal of Climatology, 27, 541–553.
Manabe, S. (1957) On the modification of air-mass over the Japan
Sea when the outburst of cold air predominates. Journal of the
Meteorological Society of Japan, 35, 311–326.
Matsumoto, S. (1967) Orographic edge effect on the downstream
cumulus activity. Journal of the Meteorological Society of Japan,
45, 500–503.
Medina, S. & Houze, R.A., Jr. (2003) Air motions and precipitation
growth in alpine storms. Quarterly Journal of the Royal Meteorological Society, 129, 345–371.
Mlawer, E.J., Taubman, S., Brown, J., Iacono, P.D. & Clough, S.A.
(1997) Radiative transfer for inhomogeneous atmospheres:
RRTM, a validated correlated-k model for the longwave. Journal of Geophysical Research, 102, 16663–16682.
Morrison, H., Thompson, G. & Tatarskii, V. (2009) Impact of cloud
microphysics on the development of trailing stratiform precipitation in a simulated squall line: comparison of one- and twomoment schemes. Monthly Weather Review, 137, 991–1007.
Nagata, M. (1991) Further numerical study on the formation of the
convergent cloud band over the Japan Sea in winter. Journal of
the Meteorological Society of Japan, 69, 419–428.

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

KUSAKA ET AL.

Nagata, M., Ikawa, M., Yoshizumi, S. & Yoshida, T. (1986) On the
formation of a convergent cloud band over the Japan Sea in
winter; numerical experiments. Journal of the Meteorological
Society of Japan, 64, 841–855.
Nakai, S., Iwanami, K., Misumi, R., Park, S.-G. & Kobayashi, T.
(2005) A classification of snow clouds by doppler radar observations at Nagaoka, Japan. Scientific Online Letters on the Atmosphere: SOLA, 1, 161–164.
Nakamura, K. & Asai, T. (1985) A numerical experiment of airmass
transformation processes over warmer sea. Part II: interaction
between small-scale convections and large-scale flow. Journal
of the Meteorological Society of Japan, 63, 805–827.
Ninomiya, K. (1964) Water-substance budget over the Japan Sea
and the Japan Islands during the period of heavy snow storm.
Journal of the Meteorological Society of Japan, 42, 317–329.
Niziol, T.A., Snyder, W.R. & Waldstreicher, J.S. (1995) Winter
weather forecasting throughout the eastern United States. Part
IV: Lake effect snow. Weather and Forecasting, 10, 61–77.
Nullet, D. & McGranaghan, M. (1988) Rainfall enhancement over
the Hawaiian islands. Journal of Climate, 1, 837–839.
Ohigashi, T. & Tsuboki, K. (2005) Structure and maintenance
process of stationary double snowbands along the coastal
region. Journal of the Meteorological Society of Japan, 83,
331–349.
Passarelli, R.E., Jr. & Braham, R.R., Jr. (1981) The role of the winter
land breeze in the formation of great Lake snow storms. Bulletin of the American Meteorological Society, 62, 482–491.
Pleim, J.E. (2007a) A combined local and nonlocal closure model
for the atmospheric boundary layer. Part I: model description
and testing. Journal of Applied Meteorology and Climatology,
46, 1383–1395.
Pleim, J.E. (2007b) A combined local and nonlocal closure model
for the atmospheric boundary layer. Part II: application and
evaluation in a mesoscale meteorological model. Journal of
Applied Meteorology and Climatology, 46, 1396–1409.
Pullen, J., Gordon, A.L., Flatau, M., Doyle, J.D., Villanoy, C. &
Cabrera, O. (2015) Multiscale influences on extreme winter
rainfall in The Philippines. Journal of Geophysical Research:
Atmospheres, 120, 3292–3309.
Puvaneswaran, K.M. & Smithson, P.A. (1991) Precipitation-levation
relationships over Sri Lanka. Theoretical and Applied Climatology, 43, 113–122.
Sawyer, J.S. (1956) The physical and dynamical problems of orographic rain. Weather, 11, 375–381.
Shimizu, H., Kawamura, R., Kawano, T. & Iizuka, S. (2017) (2017)
dynamical modulation of wintertime synoptic-scale cyclone
activity over the Japan Sea due to Changbai Mountain in the
Korean peninsula. Advances in Meteorology, 2017, 1–14.
Shinoda, Y., Kawamura, R., Kawano, T. & Shimizu, H. (2021)
Dynamical role of the Changbai Mountains and the Korean
peninsula in the wintertime quasi-stationary convergence zone
over the sea of Japan. International Journal of Climatology, 41,
E602–E615.
Siler, N. & Durran, D. (2016) What causes weak orographic
rain shadows? Insights from case studies in the cascades and
idealized simulations. Journal of the Atmospheric Sciences, 73,
4077–4099.
Sinclair, M.R., Wratt, D.S., Henderson, R.D. & Gray, W.R. (1997)
Factors affecting the distribution and spillover of precipitation
in the southern Alps of New Zealand—a case study. Journal of
Applied Meteorology, 36, 428–442.

KUSAKA ET AL.

Skamarock, W.C., Klemp, J.B., Dudhia, J., Gill, D.O., Liu, Z.,
Berner, J. et al. (2019) A description of the advanced research WRF
model version 4. Boulder, CO: NCAR Technical Note, p. 145.
Smith, R.B. & Barstad, I. (2004) A linear theory of orographic precipitation. Journal of the Atmospheric Sciences, 61, 1377–1391.
Steenburgh, W.J. & Nakai, S. (2020) Perspectives on sea-and lakeeffect precipitation from Japan's "Gosetsu Chitai". Bulletin of
the American Meteorological Society, 101, E58–E72.
Stockham, A.J., Schultz, D.M., Fairman, J.G. & Draude, A.P. (2018)
Quantifying the rain-shadow effect: results from the Peak District, British Isles. Bulletin of the American Meteorological Society, 99, 777–790.
Sunny Lim, K.S. & Hong, S.Y. (2010) Development of an effective
double-moment cloud microphysics scheme with prognostic
cloud condensation nuclei (CCN) for weather and climate
models. Monthly Weather Review, 138, 1587–1612.
Thompson, G., Field, P.R., Rasmussen, R.M. & Hall, W.D. (2008)
Explicit forecasts of winter precipitation using an improved bulk
microphysics scheme. Part II: implementation of a new snow
parameterization. Monthly Weather Review, 136, 5095–5115.
Underwood, S.J., Kaplan, M.L. & King, K.C. (2009) The role of
upstream midtropospheric circulations in the Sierra Nevada
enabling leeside (spillover) precipitation. Part I: a synoptic-scale
analysis of spillover precipitation and flooding in a leeside
basin. Journal of Hydrometeorology, 10, 1309–1326.
Veals, P.G., Steenburgh, W.J., Nakai, S. & Yamaguchi, S. (2019)
Factors affecting the inland and orographic enhancement of
sea-effect snowfall in the Hokuriku region of Japan. Monthly
Weather Review, 147, 3121–3143.
Whiteman, C.D. (2000) Mountain meteorology: fundamentals and
applications. New York: Oxford University Press, p. 355.
Yagi, S. & Uchiyama, Y. (1983) Cloud movement and confluence
around Noto peninsula and Sado Island in relation to heavy
snowfall over Johetsu area. Tenki, 30, 291–294 (in Japanese).
Yeh, H.C. & Chen, Y.L. (1998) Characteristics of rainfall distributions over Taiwan during the Taiwan area mesoscale experiment
(TAMEX). Journal of Applied Meteorology and Climatology, 37,
1457–1469.
Yoshino, M.M. (1977) The winter monsoon. In: Fukui, E. (Ed.) The
climate of Japan. Tokyo: Elsevier, pp. 65–84.
Yoshizaki, M., Kato, T., Eito, H., Hayashi, S. & Tao, W.K. (2004) An
overview of the field experiment “Winter Mesoscale Convective
Systems (MCSs) over the Japan Sea in 2001,” and comparisons
of the cold-air outbreak case (14 January) between analysis and
a non-hydrostatic cloud-resolving model. Journal of the Meteorological Society of Japan, 82, 1365–1387.

SU PP O R TI N G I N F O RMA TI O N
Additional supporting information can be found online
in the Supporting Information section at the end of this
article.
How to cite this article: Kusaka, H., Suzuki, N.,
Yabe, M., & Kobayashi, H. (2023). The
snow-shadow effect of Sado Island on Niigata City
and the coastal plain. Atmospheric Science Letters,
e1182. https://doi.org/10.1002/asl.1182

1530261x, 0, Downloaded from https://rmets.onlinelibrary.wiley.com/doi/10.1002/asl.1182 by University Of Tsukuba, Wiley Online Library on [25/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

12 of 12

この論文で使われている画像