JFA (J URNAL F ISIKA DAN A PLIKASINYA ) VOLUME 19, N UMBER 3, O CTOBER 2023

WRF-ARW Numerical Model Sensitivity Test on

Simulation of Loud Rain in The South Kalimantan Area
Abdul Hamid Al Habib*1 and Resa Agna Firdiyanto1
1
Indonesian Agency for Meteorology, Climatology and Geophysics, Kemayoran, Jakarta Pusat 10610, Indonesia
Abstract: On January 13-14, 2021, there was heavy rain in the South Kalimantan region, causing more
than 10,000 houses and the main provincial road to be flooded, and also 2 main bridges collapsed. Based on
observations at the Syamsudin Noor Meteorological Station Banjarmasin, the rainfall values on January 13 and
14 2021 were 51 mm and 249 mm, respectively. Meanwhile, at the Banjarbaru Climatology Station, it was
recorded on January 13-14, 2021, at 45.9 mm and 255.3 mm, respectively. The amount of rainfall recorded at
the Banjarmasin Meteorological Station and Banjarbaru Climatology Station makes this condition interesting to
study. This simulation uses FNL data with temporal and spatial resolution of 3 hours and 1◦ ×1◦ , respectively.
In this study, the downscaling stage was carried out 2 times with domain 1 of 16 km and domain 2 of 6 km.
Furthermore, the input data is running by testing as many as 9 parameterization schemes. Based on the results of
the WRF rainfall output with the microphysics scheme (Kessler), the PBL scheme (Yonsei University Scheme)
and the cumulus scheme (Kain-Fritcsh) showed the best value and the smallest error value compared to the
other 8 schemes. Based on the CAPE value and air humidity, it proves that the atmospheric conditions are
unstable and there is significant growth of convective clouds in the South Kalimantan region. The results of the
stremaline analysis also show the presence of strong wind bends that result in the accumulation of air masses
and indications of orographic rain in the west of the Meratus Mountains.
Keywords: WRF; Simulation; Rainfall; Parameterization.

*Corresponding author: abdulhamidalhabib96@gmail.com

Article history: Received 14 June 2023, Accepted 01 August 2023, Published October 2023.
http://dx.doi.org/10.12962/j24604682.v19i3.17421
2460-4682 c Departemen Fisika, FSAD-ITS

I. INTRODUCTION warm humidity throughout the year, high evaporation, and

high rainfall. The interaction between the sea and the land,
In kompas.com daily, from 13 to 14 January 2021, stated and the interaction between the local scale and the larger scale
that extreme weather had occurred, namely, heavy rain ac- in Indonesia make weather and climate patterns in each region
companied by lightning and strong winds which had an impact different and complex [3]. Therefore we need a mesoscale
on the flood disaster in the South Kalimantan region. Floods weather simulation method that is able to describe the ac-
occurred in most areas of the regencies of Banjarmasin, Ban- tual atmospheric conditions [4]. In this study, the WRF-ARW
jarbaru, Banjar, Tanah Laut, Barito Kuala, Tapin, Hulu Sungai model was used to simulate heavy rain.
Selatan, Hulu Sungai Utara, Balangan, and Tabalong. Based WRF-ARW (Weather Research and Forecasting - Ad-
on BPBD data, the impact of this extreme weather resulted in vanced Research WRF) is a numerical weather model that can
the submerging of more than 10,000 houses in the South Kali- simulate atmospheric conditions in an area. WRF-ARW can
mantan region. The flood disaster also caused the main roads help better assess meteorological phenomena [5]. The WRF-
of South Kalimantan Province to be inundated and two main ARW model requires parameterization to calculate processes
bridges to collapse [1]. occurring in the atmosphere. That is because the WRF-ARW
Rainfall with high intensity was recorded on January 13 model cannot fully solve the atmospheric equation explicitly
2021 of 51 mm and on January 14 2021 of 249 mm at the [6]. Microphysical and cumulus parameterization is a param-
Syamsudin Noor Meteorological Station Banjarmasin. Mean- eterization contained in the WRF-ARW model that explains
while, the Banjarbaru Climatology Station recorded 15.9 mm the process of cloud formation and rain in the model. WRF-
of rainfall on January 13, 2021, and 255.3 mm of rain on Jan- ARW model can be used to study the phenomenon of heavy
uary 14, 2021. Based on this data, the accumulated rainfall for rain.
two days at the Syamsudin Noor Meteorological Station was Different weather patterns between regions require testing
300 mm, and the accumulated rainfall at the Banjarbaru Cli- numerical weather models such as selecting parameterization
matology Station totaled 271.2 mm. According to the BMKG schemes, initial conditions, and spin-up times to produce the
Head Regulation Number 009 of 2010, a heavy rain event can best predictions [7].
be categorized as an extreme weather phenomenon [2]. The purpose of this research is to determine the best WRF-
Weather simulation around the equator is quite difficult to ARW model parameterization scheme configuration that can
do. Indonesia’s position in the tropics results in relatively be used to simulate heavy rain events in the cities of Banjar-
 sc
for January 12-14 2021 as shown in Figure 2, the estimated sc
rainfall for the WRF model from the parameterization results sc
for scheme 7 is the best and approaches the actual rainfall sc
80 compared to Habib / J.Fis.
the other dan Apl.,
8 schemes at vol. 19, no. 3, pp.
the Syamsudin 79-85,
Noor 2023
Station
in Banjarmasin and the Climatology Station in Banjarbaru. Figure
Obs rainfall and WRF rainfall Syamsudin Noor Meteorological Station Banjarmasin output of
the South
to the GS
Fritcsh)
convectiv
cloud mo
updraft a
and simp
Scheme 1 Scheme 2 Scheme 3 Scheme 4 Scheme 5 Scheme 6 Scheme 7 Scheme 8 Scheme 9

Obs rainfall and WRF rainfall at Banjarbaru Climatology Station

FIG. 1: Domain Settings using the Domain Wizard on the WPS pro-
cess
Scheme 1 Scheme 2 Scheme 3 Scheme 4 Scheme 5 Scheme 6 Scheme 7 Scheme 8 Scheme 9

masin and Banjarbaru. The best configuration of the param-

eterization scheme can be used to determine the atmospheric FIG. 2: Graph of rainfall comparison per day, on 12-14 January FIG. 3: S
conditions that occurred when heavy rains occurred in the re- 2021
FIG. between
2: Graph observation
of rainfall measurements
comparison with January
per day, on 12-14 nine the South
gion on 13-14 January 2021. The results of this study can be parameterization schemes at the Syamsudin Noor
2021 between observation measurements with nine parameterization (a) and th
Meteorological
schemes StationNoor
at the Syamsudin Banjarmasin (top)Station
Meteorological and Banjarbaru
Banjarmasin (b)
used as a reference in analyzing and predicting the potential
Climatology Station (bottom)
(top) and Banjarbaru Climatology Station (bottom)
for extreme rain hydrometeorological disasters in Indonesia in
the future. Based on table 3 below, the lowest RMSE (Root Mean Figure
Square Error) value at the Syamsudin Noor Banjarmasin (CAPE) v
Meteorological Station occurs in parameterization scheme 7 region, e
from microphysical settings, another parameterization scheme
that exp
II. METHODOLOGY is with a value of
the default 12.950,namely
setting, which means that the
the Dudhi schemefor
scheme canthe
predict
long- Energy (C
the amount
wave radiationof rainfall
schemewith the smallest
based difference
on simple in error
efficient to the
integration
amount of observed rainfall compared to the other eight certain di
This research was carried out in several work steps, namely applied to absorption and scattering of sunny and cloudy. indicate t
schemes. As for the Banjarbaru Climatology Station, the lowest
identifying the time and place of flooding in the South The YSUvalue
RMSE Boundary
is shownLayer (PBL) 7scheme
by scheme with a reduces
value of non-local
105.838 stated tha
Kalimantan region, identifying rain cloud growth using the mixing
where effects
this valueand includes
shows explicit
the smallest entrainment
value. Based on thefluxes
RMSE,of CAPE va
Himawari-8 satellite, running rainfall prediction simulations heat, moisture, and momentum, as well as differences
shows that scheme 7 is the best scheme that can predict in PBL
the identify t
using the WRF-ARW model with 9 (nine) selected scheme amount of rainfall
specification heights.in the South Kalimantan
Kain-Fritsch Cumulus region.
parameterization convectiv
configurations, verifying the results and analysis. scheme (new Eta) based on CAPE calculations, Grell Devenyi
The data used as input data for the WRF-ARW model is scheme, Betts Miller Janjic scheme based on convective ad-
FNL (Final Analysis) data from 12 January 2021 at 00 UTC to justment including deep and shallow convection.
14 January 2021 at 00.00 UTC. This FNL data can be accessed The Kessler scheme is a simple warm cloud schematic and
at http://rda.ucar.edu with a spatial resolution of 1◦ ×1◦ and a includes water vapor, clouds, and rain. The microphysical
temporal resolution of 6 hours. The length of data used in the processes included are: the production, falling, and evapo-
event simulation is 48 hours, with 24 hours for the spin-up ration of rain, accretion and autoconversion of cloud water,
time and a further 30 hours for the event simulation analysis and production of cloud water from condensation [8]. YSU’s
period. PBL (Yonsei University Scheme) scheme is a Planet Bound-
In the WPS (WRF Preprocessing System) process, the ary Layer scheme developed by the staff of the Department of
WRF-ARW model uses the WRF Domain Wizard to make it Atmospheric Sciences at Yonsei University to produce a more
easier to define domains, create name lists automatically, and realistic model of the PBL structure and its development [9].
run WPS (geogrid.exe, ungrib.exe, and metgrid.exe) automat- The KF scheme is mass flux parameterized and uses the La-
ically. So that all the output from the WRF Domain Wizard grangian parcel method, and in general, it can be grouped into
can be directly used to run the WRF model. Domain 1 (D01) three parts: 1) convective trigger function, 2) mass flux for-
is located between 106.51-123.88 E and 10.26 LS-3.68 LU mulation, and 3) closing assumptions. The initial version of
with a resolution of 18 km, domain 2 (D02) is located between the KF scheme used a simple cloud model with updraft and
112.19-117.17 E and 5.62 LS-1.09 LS with a resolution of 6 downdraft humidity and was modified for use by the NWP
km. The center point of the domain in this study is at the (Numerical Weather Prediction) model [10]. The physical op-
Syamsudin Noor Banjarmasin Meteorological Station, which tions in the configuration settings for the WRF model scheme
is at 3.437 South Latitude and 114.75 East Longitude Fig. 1. in this study are shown in Table I below. The experimental
Then the running WRF process (real.exe and wrf.exe) is names of the various WRF model parameterization schemes
carried out with the configuration as shown in Table I. Apart tested in this study are shown in Table II.
 convective cloud growth to occur. This scheme uses a simple
cloud model that takes into account the equations of mass flux,
updraft and downdraft, humidity, entrainment and detrainment
Habib / J.Fis. dan Apl., vol. 19, no. 3, pp. 79-85, 2023 and simple microphysical processes [11]. 81
Scheme 1 Scheme 2 Scheme 3 Scheme 4 Scheme 5 Scheme 6 Scheme 7 Scheme 8 Scheme 9

TABLE I: WRF Model Parameterization Scheme Configuration

Arrangement Information
Obs rainfall and WRF rainfall at Banjarbaru Climatology Station
Running WRF Domain 1 Domain 2
history interval 180 sec 60 sec
time step 60 sec 60 sec
e we 100 88
e sn 91 79

ra sw physics Dudhi scheme Dudhi scheme

Scheme 1 Scheme 2 Scheme 3 Scheme 4 Scheme 5 Scheme 6 Scheme 7 Scheme 8 Scheme 9 (a) (b)
bl pbl physics YSU scheme YSU scheme

Kain Fritsch (new Kain Fritsch (new

FIG. 2: Graph of rainfall comparison per day, on 12-14 January FIG. 3: Spatial analysis of rainfall on 12 – 14 January 2021 for
cu physics Eta) scheme Eta) scheme
2021 between observation measurements with nine theFIG.
South Kalimantan
3: Spatial analysisarea of theon7 12
of rainfall WRF-ARW
14 January model
2021 forscheme
the
Betts Miller Janjic Betts Miller Janjic
parameterization schemes at the Syamsudin Noor (a)South
and the accumulation
Kalimantan area of of
theGSMAP
7 WRF-ARWfrommodel
the Himawari
scheme (a)satellite
and
Grell Devenyi Grell Devenyi
Meteorological Station Banjarmasin (top) and Banjarbaru (b)the accumulation of GSMAP from the Himawari satellite (b)
Climatology Station (bottom) WRF WRF
mp physics Single-Moment 6 Single-Moment 6
CAPE (J/Kg) (00UTC 12Jan2021) CAPE (J/Kg) (12UTC 12Jan2021)
Class (WSM6) Class (WSM6)
Based on table 3 below, theKessler
lowest RMSE (Root Mean Figure 4 shows the Convective Available Potential Energy
Kessler (CAPE) values on 12-14 January 2021 in the South Kalimantan
Square Error) value at the Syamsudin Noor Banjarmasin
Perduelin Perduelin
Meteorological Station occurs in parameterization scheme 7 region, especially to the west of the Meratus mountains (areas
with a value of 12.950, which means that the scheme can predict that experienced flooding). Convective Available Potential
the amount of rainfall with the smallest difference in error to the Energy (CAPE) is the amount of energy a parcel of air can lift a
amount of observed rainfall compared to the other eight certain distance vertically in the atmosphere [12]. CAPE can also
schemes. As for the Banjarbaru Climatology Station, the lowest indicateCAPE that(J/Kg)
the atmosphere is in anCAPE
(00UTC 13Jan2021) unstable
(J/Kg) condition [13]. ItFIG.
(12UTC 13Jan2021) is 6: A
RMSE valueII:isExperiment
TABLE shown byName
scheme
WRF 7Model
withParameterization
a value of 105.838
Scheme stated that the air column will experience strong convection if the
2021 for
where this value shows the smallest
Experiment value. Based
Parameterization on the RMSE,
Scheme CAPE value is greater [14]. The CAPE value is well used to Figure
shows thatName
scheme 7Microphysics
is the best scheme
PBL Cumulusthe identify the level of atmospheric stability and the potential for
that can predict South Ka
amount of rainfall
Scheme 1 in the South
WRF Kalimantan
Yonseiregion. Kain Fritcsh convective activity. that the a
with valu
Scheme 2 Single-Moment University Betts Miller in the Sou
6-Class Scheme (YSU Janjic C precise
(WSM6) Scheme)
Scheme 3 Grell Devenyi CAPE (J/Kg) (00UTC 14Jan2021) CAPE (J/Kg) (12UTC 14Jan2021) western s
Scheme 4 Yonsei Kain Fritcsh air tempe
Scheme 5 Kessler University Betts Miller the wind
Scheme (YSU Janjic to the ea
Scheme) warmer th
Scheme 6 Grell Devenyi
Scheme 7 Yonsei Kain Fritcsh (00
Scheme 8 Perduelin University Betts Miller
Scheme (YSU Janjic
Scheme 9 Scheme) Grell Devenyi

FIG. 4: Cape values from January 12 to 14, 2021 1000 mb

layer
FIG. 4: toCape
800 values
mb layer
from January 12 to 14, 2021 1000 mb layer to (12
800 mb layer

Based on the CAPE map in Figure 4 above. The CAPE value

Verification of the simulation results from each WRF-ARW of the surface layer up to 925 mb ranges from 900 – 2700 J/kg.
model scheme used rainfall observation data from the Syam- The highest CAPE III. RESULT
value was AND DISCUSSION
over 2700 J/Kg in the 1000 – 950
sudin Noor Meteorological Station in Banjarmasin and the mb layer on January 13, 2021 at 00 UTC. This value indicates
Climatology Station in Banjarbaru. Spatial verification uses aFig.
very 2unstable level of atmospheric
is a comparison diagram betweenstabilityobswithrainfall
strong at
daily rainfall data from the GSMAP (Global Satellite Map- the Syamsudin Noor Meteorological Station Banjarmasinthat
intensity. The unstable atmospheric conditions indicate and
ping of Precipitation) satellite which can be accessed from there is potential
Banjarbaru for convective
Climatology Stationactivity to occur
with rainfall from the FIG. 7: L
and produce
output
heavy rains of quite a long duration. This is similar to the results 12 to 14,
the site ftp://hokusai.eorc.jaxa.jp. The GSMaP data used is WRF model. Based on the daily rainfall values for January 12-
of research conducted in the Bogor area which shows that
GSMaP NRT daily data, which contains real-time rain rates 14between
2021 asvariations
shown in Fig. 2, the estimated rainfall for the WRF Water
in the atmospheric stability index and formation
(mm/hour) as long as the data format is *.DAT.GZ. The model fromrainfall
increased the parameterization results for as
are directly proportional, scheme
well as 7 isthethe
vapor tran
daily rain intensity rain rate data has a spatial resolution of best and approaches the actual rainfall compared
growth of cumulonimbus clouds during heavy rain events [15]. to the other
moisture
0.25◦ ×0.25◦ . 8 schemes at the Syamsudin Noor Station in Banjarmasin and from nort
Banjarmasin (00z12Jan-00z15Jan2021) Banjarbaru (00z12Jan-00z15Jan2021)
pressure l
the surfac
vapor is o
The r
January 1
show that
west. Bas
mb layer,
(a) (b) there was
kg/ms-1)
FIG. 5: Relative Air Humidity (%) layer 1000 – 100 mb from
82 Habib / J.Fis. dan Apl., vol. 19, no. 3, pp. 79-85, 2023

Banjarmasin (00z12Jan-00z15Jan2021) Banjarbaru (00z12Jan-00z15Jan2021)

TABLE III: Statistical Error (RMSE) values for nine output rainfall
parameterization schemes for the WRF model with obs rain gauges
at the Syamsudin noor Banjarmasin Meteorological Station and Ban-
jarbaru Climatological Station
Experiment Name Banjarmasin Banjarbaru
scheme 1 64.835 115.274
scheme 2 127.191 145.904
(a) (b)
scheme 3 115.038 128.937
scheme 4 107.099 132.334
FIG. 5: Relative Air Humidity (%) layer 1000-100 mb from January scheme 5 128.334 144.864
12 to 14, 2021 at Banjarmasin (a), and Banjarbaru (b) scheme 6 125.294 143.964
scheme 7 12.950 105.838
scheme 8 113.713 135.870
the Climatology Station in Banjarbaru. scheme 9 113.745 129.751
Based on Table III, the lowest RMSE (Root Mean Square
Error) value at the Syamsudin Noor Banjarmasin Meteoro-
logical Station occurs in parameterization scheme 7 with a Air humidity is the amount of water vapor contained in the
value of 12.950, which means that the scheme can predict air or atmosphere [16]. It can be seen from the surface layer
the amount of rainfall with the smallest difference in error to (1000-800 mb) before and during the rain event that the air
the amount of observed rainfall compared to the other eight condition in the Banjarmasin and Banjarbaru areas is quite
schemes. As for the Banjarbaru Climatology Station, the saturated with a value of 80 100%. In the middle layer (700-
lowest RMSE value is shown by scheme 7 with a value of 500 mb) 80-100% and in the top layer (500 200 mb) the air
105.838 where this value shows the smallest value. Based on humidity is still very saturated in the range of 60-100%. This
the RMSE, shows that scheme 7 is the best scheme that can indicates that the air condition is very humid during periods
predict the amount of rainfall in the South Kalimantan region. of heavy rain and supports the condensation process in the
Fig. 3 is estimated rainfall accumulation data from the out- formation of rain clouds.
put of the parameterization scheme of the 7 WRF models in Fig. 6 shows the air temperature conditions in the South
the South Kalimantan region on January 12-14 2021 com- Kalimantan region on January 12-14 2021. It can be seen
pared to the GSMAP satellite. Scheme 7 uses the cumulus that the air temperature in the South Kalimantan region varies
KF (Kain-Fritcsh) scheme which is used to predict the condi- with values ranging from 20-27 ◦ C. The lowest air tempera-
tions for convective cloud growth to occur. This scheme uses ture in the South Kalimantan region on January 12-14 2021
a simple cloud model that takes into account the equations of worth 20 ◦ C precisely at the top of the Meratus mountain
mass flux, updraft and downdraft, humidity, entrainment and range. The western slopes (leeward) of the Meratus moun-
detrainment and simple microphysical processes [11]. tains have hotter air temperatures due to adiabatic processes
Fig. 4 shows the Convective Available Potential Energy than the slopes above the wind (windward), and when return-
(CAPE) values on 12-14 January 2021 in the South Kaliman- ing to the ground surface to the east of the Meratus mountains
tan region, especially to the west of the Meratus mountains the temperature returns warmer than when the air has not yet
(areas that experienced flooding). Convective Available Po- climbed the mountains [17].
tential Energy (CAPE) is the amount of energy a parcel of air Water vapor transport is one of the main components in the
can lift a certain distance vertically in the atmosphere [12]. formation of rain clouds. Large air masses and changes in
CAPE can also indicate that the atmosphere is in an unstable water vapor transport can shift places where rain forms. The
condition [13]. It is stated that the air column will experi- process of moisture transport at the equator occurs when water
ence strong convection if the CAPE value is greater [14]. The vapor moves from north to south or vice versa. According to
CAPE value is well used to identify the level of atmospheric Zhou (2003) the pressure limit used in the water vapor trans-
stability and the potential for convective activity. port equation is from the surface to a height of 300 hPa [18].
Based on the CAPE map in Fig. 4. The CAPE value of the This is because water vapor is only concentrated in the lower
surface layer up to 925 mb ranges from 900-2700 J/kg. The troposphere.
highest CAPE value was over 2700 J/Kg in the 1000-950 mb The results of the analysis of water vapor transport from
layer on January 13, 2021 at 00 UTC. This value indicates a January 12, 2021, at 00 UTC to January 14, 2021, at 12
very unstable level of atmospheric stability with strong inten- UTC show that there is a movement of water vapor transport
sity. The unstable atmospheric conditions indicate that there from the west. Based on the water vapor transport map in
is potential for convective activity to occur and produce heavy the 925-300 mb layer, on January 12, 2021, at 00 UTC it be-
rains of quite a long duration. This is similar to the results gan to be seen that there was a fairly high intensity of water
of research conducted in the Bogor area which shows that vapor (1000-1200 kg/ms-1) in the Meratus mountain range
between variations in the atmospheric stability index and in- (Fig. 7). On January 12, 2021, at 12 UTC in the southern
creased rainfall are directly proportional, as well as the growth region of South Kalimantan Province, an increase in water
of cumulonimbus clouds during heavy rain events [15]. vapor intensity of more than 1200 kg/ms-1 was evenly seen.
Habib / J.Fis. dan Apl., vol. 19, no. 3, pp. 79-85, 2023 83
g
(00UTC 12Jan2021) (00UTC 13Jan2021) (00UTC 14Jan2021)
(00UTC 12Jan2021) (00UTC 13Jan2021) (00UTC 14Jan2021)
T
2
in
(00UTC 12Jan2021) C
a
0
th
s
FIG. 6: Air temperature (◦ C) on 00 UTC from January 12 to 14, 1
2021 for South Kalimantan -
FIG. 8: Streamline layer of 3000 feet January 12-14 2021, South
-
Kalimantan area
FIG. 8: Streamline layer of 3000 feet January 12-14 2021, South
(00UTC 12Jan2021) (12UTC 12Jan2021) (00UTC 13Jan2021) Kalimantan area J
Based on figure 8, from 12 to 14 January, the wind blew a
from the southwest-west with an average speed of 2-3 knots. c
Thatmantan
confirms regionthat in January,
causes thecarries
the wind that Souththe Kalimantan
water vapor region
to
enters
movetheuprainy season (the
the mountains, climbdominant
the slopeswind
and getis higher
from the(Fig.west
8). The
(Asian higher the
monsoon) thatwater vapor
carries carried
wet by the wind,
air). Based on thethestreamlined
higher
the potential for condensation to occur. This is because the
map, from 12 to 14 January 2021, there was a disturbance in the
(12UTC 13Jan2021) (00UTC 14Jan2021) (12UTC 14Jan2021) higher the upper layer, the colder the air temperature (Fig. 8).
formThis
of acondensation
shear line in or the South Kalimantan
condensation will form rainregion.
cloud seeds s
The
or water droplets. After experiencing saturation, inthethe
existence of the Meratus Mountains South
water K
Kalimantan region causes the wind that carries the
droplets that are above will reduce the water content in them. water vapor d
to move up thewater
These falling mountains, climb
droplets are theorographic
called slopes and rain get
[17].higher c
(figureBased
8). The higher
on Fig. 9, the water vapor
an analysis of thecarried by the
time series of wind,
cloud the p
higher the from
growth potential for condensation
the infrared to occur.
channel Himawari Thisimagery.
satellite is because s
The process
the higher of convective
the upper layer, thecloud growth
colder thebegins on January 13,
air temperature (figure p
6). This condensation or condensation will form ofrain
2021, around 03.30 UTC, marked by the beginning a de-
cloud S
crease in the cloud top temperature in Banjarmasin City to -
seeds or water droplets. After experiencing saturation, the water
68 ◦ C. Convective activity continued until around 06.00 UTC R
FIG. 7: Layer water vapor transport of 925-300 mb from January 12 droplets that an
there was areextreme
abovedecrease
will reduce
in thethe water
cloud content in to
top temperature them. T
to 14, 2021 for the South Kalimantan region These falling
◦ water droplets are called orographic
-75 C. At 09.30 UTC the cloud top temperature rose to -33.3 rain [17]. s
◦
C, indicating that at that time the dissipation phase began. h
-112.5
After that, on the same day, the clouds experienced a growth
The increase in the intensity of water vapor is increasingly -100.0 phase again at around 13.30 UTC marked by a decrease in the a
widespread throughout the South Kalimantan region until Jan- -80.0
cloud top temperature to -68 ◦ C. The cloud top temperature c
uary 14, 2021, at 12 UTC. -60.0
increased from 18.30 UTC to -40 ◦ C, then the cloud top tem-
The large water vapor transport value indicates the poten- -40.0
perature remained constant until January 14, 2021, at 06.00
a
tial for rain clouds to cause flooding in the South Kaliman- -20.0
UTC. This indicates that the clouds are dominated by convec- c
tan region. That is like the results of previous studies which 0.0
tive clouds (Cumulonimbus) and rain clouds (Nimbostratus) r
state that water vapor has an important role in predicting fu- 20.0
throughout the day. in
ture rainfall and large amounts of water vapor can cause high 40.0
r
o
60.0
rainfall [19, 20].
80.0
Streamline wind is a condition of wind currents that blow 97.5
IV. CONCLUSION C
which is drawn based on a tangent or parallel to the wind data im
in the area they are located [21]. Fig. 8 shows streamline In general, the WRF-ARW model was quite good at simu- w
analysis is carried out from an altitude of 3000 feet to the top FIG.lating
9: Cloud top temperature time series (himawari satellite)
the phenomenon of heavy rain in the South Kalimantan
of the atmosphere (925mb layer and above) to see the actual from January 12 to 14,2021,
2021 at was
the demonstrated
Syamsudin byNoor h
region on January 12-14 which
wind without the influence of frictional forces and the influ- Meteorological
the WRF-ARW Station Banjarmasin
model’s ability to identify convective cells.
ence of topography. The best parameterization scheme that can predict rainfall
Based on Fig. 8, from 12 to 14 January, the wind blew from in the South Kalimantan region is shown by scheme 7 with
the southwest-west with an average speed of 2-3 knots. That the smallest RMSE value compared to other parameteriza-
confirms that in January, the South Kalimantan region enters tion schemes with a value of 12,950 at the Syamsudin Noor
the rainy season (the dominant wind is from the west (Asian Banjarmasin Meteorological Station and an RMSE value of
monsoon) that carries wet air). Based on the streamlined map, [1]105,838Harian atKompas. “BMKG:
the Banjarbaru Banjir Kalimantan
Climatology Station. TheSelatan Akibat
best pa-
from 12 to 14 January 2021, there was a disturbance in the Cuaca Ekstrem
rameterization Dipicu
scheme Dinamika
for the WRF modelAtmosfer
in this Labil,” Internet:
study can be
form of a shear line in the South Kalimantan region. https://www.kompas.com/sains/read/2021/01/18/120500823/b
used as a reference in operational forecasting of heavy rains
The existence of the Meratus Mountains in the South Kali- inmkg--banjir-kalimantan-selatan-akibat-cuaca-ekstrem-dipicu-
the South Kalimantan region.
dinamika?page=all, accessed 25 November 2021.
[2] Peraturan Kepala BMKG Nomor. KEP 009 Tahun 2010.
“Prosedur Standar Operasional Pelaksanaan Peringatan Dini,
Pelaporan, dan Diseminasi Informasi Cuaca Ekstrim”, BMKG,
2010.
[3] Qian, J. H. “Why precipitation is mostly concentrated over
islands in the Maritime.” American Meteorological Society
Journals, 2008.
[4] Rizkiana, D., Josephine., Syahidah, M., Amelda, P., Arida, V.
“Perbandingan Skema Parameterisasi Dalam Simulasi Cuaca
 droplets that are above will reduce the water content in them. The
84 These falling water droplets are called Habib orographic
/ J.Fis. dan Apl., rain [17].
vol. 19, no. 3, pp. 79-85, 2023 study

heav
-112.5 B
-100.0

-80.0
atmo
-60.0
conv
-40.0 amou
-20.0 cloud
0.0
resul
20.0
indic
40.0
resul
orog
60.0

80.0

97.5 Com
imag
FIG. 9: Cloud top temperature time series (himawari satellite) with
FIG. 9: Cloud top temperature time series (himawari satellite) from January 12 to 14, 2021 at the Syamsudin Noor Meteorological Station
Banjarmasin

from January 12 to 14, 2021 at the Syamsudin Noor heav

Meteorological
the atmospheric Station
conditions are unstable and thereBanjarmasin
Based on the CAPE value and air humidity, it is clear that
is signifi-
bend) which results in the accumulation of air masses and in-
dications of orographic rain occurring to the west of the Mer-
cant convective cloud growth in the South Kalimantan region. atus Mountains. Comparing the results of the analysis with
A large amount of water vapor transport indicates the poten- Himawari satellite imagery clarifies the conditions that occur
tial for rain clouds to cause flooding in the South Kalimantan during heavy rains with the lowest cloud top temperature of
region. The results of the analysis of airflow (streamline) and -75 ◦ C indicating very heavy rain that occurred.
temperature indicate the existence of a shear line (strong wind

[1] Harian Kompas. “BMKG: Banjir Kalimantan

[1] Harian Kompas, ”BMKG: Banjir Kalimantan Selatan Akibat
erasional SelatanDengan
Di Jabodetabek: Perbandingan AkibatModel Wrf,”
Cuaca Ekstrem Dipicu Dinamika Atmosfer Labil,” Internet:
Jurnal Meteorologi Dan Geofisika Vol. 13 No. 2 Tahun 2012.
Cuaca Ekstrem Dipicu Dinamika Jakarta:Atmosfer
BMKG, 2012. Labil,” Internet:
https://www.kompas.com/sains/read/2021/01/18/120500823/bmkg–

https://www.kompas.com/sains/read/2021/01/18/120500823/b
banjir-kalimantan-selatan-akibat-cuaca-ekstrem-dipicu- [8]
L. J. Wicker and R. B. Wilhelmson, ”Simulation and analysis
dinamika?page=all, accessed 25 November 2021.
of tornado development and decay within a three-dimensional
mkg--banjir-kalimantan-selatan-akibat-cuaca-ekstrem-dipicu-
[2] Peraturan Kepala BMKG Nomor. KEP 009 Tahun 2010.
supercell thunderstorm,” Journal of the atmospheric sciences,
”Prosedur Standar Operasional Pelaksanaan Peringatan Dini,
52(15), pp.2675-2703, 1995. [6]
2010.
dinamika?page=all, accessed 25 sion
November
Pelaporan, dan Diseminasi Informasi Cuaca Ekstrim,” BMKG,
S. Y. Hong, Y. Noh,2021.
[9]
and J. Dudhia, ”A New Vertical
Package With An Explicit Treatment Of Entrainment Pro-
Diffu-

[2] Peraturan Kepala BMKG Nomor. KEPweather

[3] Qian, J. H, ”Why precipitation is mostly concentrated over is-
cesses,” Monthly
lands in the Maritime,” American Meteorological Society Jour-
009
[10]
review.Tahun 2010. 2006. [7]
134(9), pp. 2318-2341,
J. S. Kain and J. M. Fritsch, ”A One-Dimensional Entrain-
“Prosedur Standar Operasional ing/Detraining
nals, 2008. Pelaksanaan Plume Model Peringatan
And Its ApplicationDini,
In Convective
[4] D. Rizkiana, et al., ”Perbandingan Skema Parameterisasi Dalam
Parameterization,” Journal of Atmospheric Sciences, 47(23),
Pelaporan, dan Diseminasi Informasi
Simulasi Cuaca Numerik Menggunakan Model WRF-ARW Cuaca
pp. 2784-2802, 1990. Ekstrim”, BMKG,

2010.
(Studi Kasus Hujan Ekstrim di Balikpapan Tanggal 5 Juli [11]
S. Ginting, et al., ”Pengaruh Parameterisasi Kumulus terhadap
2008),”, Program Studi Meteorologi, Institut Teknologi Ban-
Simulasi Angin Kencang di Makassar dengan Menggunakan
[3] Qian, J. H. “Why precipitationK.
dung, Bandung, 2011.
is mostly concentrated over
WRF,” Institut Teknologi Bandung, Bandung, 2011.
[5] G. Sulung, M. Priyanka, and N. Saraswati. ”Pengaruh Pa- [12]
Husna and M. A. Munandar, ”Analisis Nilai Convective
[8]
islands in the Maritime.” American
rameterisasi Kumulus Terhadap Simulasi Angin Kencang DiMeteorological
Available Potential
Makasar Dengan Menggunakan WRF,” Institut Tekonologi
Energy (CAPE) SelamaSociety
Tahun 2013-2016
Terhadap Hujan di Jakarta.” Prosiding Seminar Nasional Sains
Journals, 2008.
Bandung, Bandung, 2011. dan Teknologi Fakultas Teknik Universitas Wahid Hasyim Se-
[6] COMET Program, 2006. Internet:
marang, vol. 1, pp. 3033, 2017,
[4] Rizkiana, D., Josephine., Syahidah, M.,et al,Amelda,
https://www.meted.ucar.edu/nwp/, accessed 19 November
M. Alfan, ”Analisis Profil P.,
[13] CAPEArida,
(Convective V.Available

Pengkajian [9]
2021. Potential Energy) Radiometer Selama Kegiatan Intensive Ob-
“Perbandingan Skema Parameterisasi Dalam
[7] I. Gustari, et al., ”Akurasi Prediksi Curah Hujan Harian Op-
servation Period Simulasi
(IOP) di Dramaga Cuaca
Bogor,” Badan
Numerik Menggunakan Model WRF – ARW (Studi Kasus Hujan
Ekstrim di Balikpapan Tanggal 5 Juli 2008).”, Program Studi
Meteorologi, Institut Teknologi Bandung, Bandung, 2011. [10]
[5] G. Sulung, M. Priyanka, N. Saraswati. “Pengaruh Parameterisasi
Habib / J.Fis. dan Apl., vol. 19, no. 3, pp. 79-85, 2023 85

dan Penerapa Teknologi, Serpong, 2016. (BMKG) Jakarta, 2006.

[14] B. A. Kusumo, et al. ”Simulasi fenomen Squall Line Sumatra [18] T.-J. Zhou, ”Comparison of the global air-sea freshwater ex-
dengan Model WRF (Studi kasus pada tanggal 30 April 2007),” change evaluated from independent data sets.” Prog. Natural
Institut Teknologi Bandung, Bandung, 2011. Sci., 13(8), pp.626-631, 2003.
[15] L. N. Fu’adah, et al., ”Kajian Indeks Stabilitas Atmoster Ter- [19] A. Mansyur, ”Penentuan Kandungan Uap Air Mampu Curah
hadap Kejadian Hujan Lebat Di Wilayah Bogor,” Prosiding Dengan Menggunakan Terra/Aqua MODIS,” Depok, Universi-
SNFA (Seminar Nasional Fisika dan Aplikasinya, vol. 3, pp. tas Indonesia, 2010.
163, 2018. [20] [20] T. Nuraya, ”Analisis Hujan Ekstrim Berdasarkan Param-
[16] Y. S. Swarinoto and Sugiyono, ”Pemanfaatan Suhu Udara dan eter Angin dan Uap Air di Kototabang Sumatera Barat,” Pon-
Kelembaban Udara dalam Persamaan Regresi untuk Simulasi tianak, Universitas Tanjungpura, 2016.
Prediksi Total Hujan Bulanan di Bandar Lampung,” Jurnal Me- [21] P. A. Winarso, ”Analisa Cuaca I, Akademi Meteorologi dan Ge-
teorologi dan Geofisika, 12(3), pp.271-281, 2011. ofisika (AMG),” Jakarta, 2011.
[17] B. Tjasyono, ”Meteorologi Indonesia I, Karakteristik & Sirku-
lasi Atmosfer,” Badan Meteorologi Klimatologi dan Geofisika