Quality Engineering,
Journal Year:
2023,
Volume and Issue:
36(3), P. 575 - 593
Published: Sept. 11, 2023
AbstractThis
article
presents
a
Statistical
Process
Control
(SPC)
framework
considering
the
response
process
as
unit
variable,
which
demands
special
treatment.
This
study
designed
Shiny
app
related
to
data
visualization
and
inferential
estimation
adopting
SPC
charts
Extreme
Value
Theory.
We
also
proposed
new
flexible
probabilistic
model
(named
FlexShape),
is
simple
yet
overcomes
skew
information
bimodality
in
historical
data,
part
of
complex
learning
task.
Results
showed
that
enables
it
handle
sets.
As
an
example,
we
presented
storytelling
from
water
particle
monitoring
(relative
humidity)
one
Atacama
Desert
station,
known
be
driest
areas
on
Earth,
across
hidden
patterns
such
inundation
microweather.
Finally,
developed
makes
possible
any
research
univariate
decision-making,
enabling
database
import
adjusting
some
parametric
models,
comparison
different
units'
distribution
goodness-of-fit.Keywords:
asymmetry
databimodal
distributioniterative
analysisrates
proportions
monitoringR
shiny
Disclosure
statementNo
potential
conflict
interest
was
reported
by
authors.Additional
informationFundingThis
funded
Universidad
de
grant
number
ATA1956
–
CC88433.
partially
supported
Vicerrectoría
Investigación
y
Postgrado
(VRIP)
Dirección
(UDA).
The
author
David
Elal-Olivero
DIUDA
REGULAR
project
No.
22409
Atacama,
Chile.
Paulo
H.
Ferreira
acknowledges
support
Brazilian
National
Council
for
Scientific
Technological
Development
[CNPq,
307221/2022-9].Notes
contributorsDiego
C.
NascimentoDiego
Nascimento
Associate
Professor
at
Copiapó,
He
holds
Ph.D.
degree
Statistics
Federal
University
São
Carlos/University
(UFSCar/USP),
M.Sc.
Business
Management
Pernambuco
(UFPE),
B.Sc.
Rio
Grande
do
Norte
(UFRN).
works
mainly
following
topics:
statistical
learning,
analytics.Oilson
A.
Gonzatto
JuniorOilson
Junior
(USP),
Carlos,
Paulo,
Brazil.
received
his
2021
UFSCar/USP,
M.Sc
Biostatistics
2017
B.Sc
2016
both
State
Maringá
(UEM),
Maringá,
Paraná,
Brazil,
licentiate
Mathematics
2014
Paraná
(UNESPAR).
has
Postdoctoral
training
2021–2023.
Currently
researches
survival
reliability
analysis.David
Elal-OliveroDavid
Full
Ciencias
Matemáticas
1987
Complutense
Madrid,
Spain.
His
main
interests
include
theory.Estefania
BonnailEstefania
Bonnail
She
her
Marine
Coastal
(Erasmus
Mundus
program)
Cádiz,
done
intensive
field
ecotoxicology.Paulo
FerreiraPaulo
Institute
Statistics,
Bahia
(UFBA),
Ph.D.,
degrees
all
Carlos
(UFSCar),
analysis,
mining
control.
npj Climate and Atmospheric Science,
Journal Year:
2025,
Volume and Issue:
8(1)
Published: Feb. 17, 2025
Global
warming
is
accelerating
climate
disasters
by
triggering
tipping
points
in
various
Earth
systems.
Although
changes
precipitation
patterns
High-Mountain
Asia
(HMA)
have
been
extensively
studied,
the
specific
thresholds
that
trigger
rapid
snowfall
loss
remain
unclear.
A
continuous
piecewise
linear
regression
model
was
employed
to
classify
HMA
into
four
distinct
regimes:
insensitive
snowfall-dominated
areas,
sensitive
rainfall-dominated
and
areas.
Our
results
show
future
will
increase
sensitivity
of
winter
spring
change,
whereas
summer
autumn
become
less
sensitive.
All
regimes
exhibit
an
upward
shift
higher
elevations,
with
varying
rates
elevation
gain
across
regions
seasons.
Temperature
primary
driver
loss,
relative
humidity
mitigates
it.
This
study
identifies
high-risk
areas
vulnerable
help
guide
development
effective
mitigation
strategies.
Water,
Journal Year:
2022,
Volume and Issue:
14(7), P. 1122 - 1122
Published: March 31, 2022
The
precipitation
phase
(i.e.,
rain
and
snow)
is
important
for
the
global
hydrologic
cycle
climate
system.
objective
of
this
study
to
evaluate
precipitation-phase
partitioning
capabilities
remote
sensing
reanalysis
modeling
methods
on
scale.
Specifically,
observation
data
from
National
Centers
Environmental
Prediction
(NCEP)
Automated
Data
Processing
(ADP),
2000
2007,
are
used
rain–snow
discrimination
accuracy
Integrated
Multi-Satellite
Retrievals
Global
Precipitation
Measurement
(IMERG)
fifth-generation
product
European
Centre
Medium
Range
Weather
Forecasts
(ERA5).
results
show
that:
(1)
ERA5
performs
better
than
IMERG
at
distinguishing
rainfall
snowfall
events,
overall.
(2)
has
high
in
all
continents
except
South
America,
while
well
only
Antarctica
North
America.
(3)
Compared
with
IMERG,
can
more
effectively
capture
events
latitudes
but
shows
worse
performance
mid-low
latitude
regions.
Both
have
lower
under
heavy
precipitation.
Overall,
provide
references
application
improvement
products.
Journal of Hydrology Regional Studies,
Journal Year:
2024,
Volume and Issue:
53, P. 101757 - 101757
Published: April 3, 2024
Precipitation
phase
in
China
has
changed
with
global
warming,
and
accurate
partitioning
of
the
precipitation
is
crucial
for
understanding
hydrological
processes
energy
balance.
Based
on
a
36-year
daily
meteorological
dataset
from
Meteorological
Administration,
this
study
conducted
comprehensive
evaluation
six
empirical
methods
estimating
nationwide
explored
applicability
distinction
across
different
meteorological,
topographic
geographic
categories.
These
utilized
inputs
air
temperature,
two
variants
incorporating
relative
humidity,
encompassed
range
combined
non-linear
temperature
hydrometeor
to
more
simple
threshold
methods.
Methods
implementing
functions
without
thresholds
performed
better
than
those
relying
solely
thresholds.
The
exponential
humidity
exhibited
best
performance,
while
only
worst
overall
performance.
Optimal
were
identified
recommended
under
scenarios
(i.e.,
multivariate
parameters,
regions).
rain–precipitation
ratio
significant
increasing
trend,
stations
showing
an
increase
primarily
concentrated
three
major
snow-covered
areas,
west
vicinity
"Hu
Huanyong
Line".
This
can
be
applied
other
regions
offers
valuable
insights
analyzing
data
developing
models.
Remote Sensing,
Journal Year:
2025,
Volume and Issue:
17(6), P. 945 - 945
Published: March 7, 2025
Winter
mixed-phase
precipitation
(P)
impacts
transportation,
electric
power
grids,
and
homes.
Forecasting
winter
such
as
freezing
(ZP),
rain
(ZR),
drizzle
(ZL),
ice
pellets
(IPs),
the
snow
(S)
(R)
boundary
remains
challenging
due
to
complex
cloud
microphysical
dynamical
processes
involved,
which
are
difficult
predict
with
current
numerical
weather
prediction
(NWP)
models.
Understanding
these
based
on
observations
is
crucial
for
improving
NWP
To
aid
this
effort,
Environment
Climate
Change
Canada
deployed
specialized
instruments
Vaisala
FD71P
OTT
PARSIVEL
disdrometers,
measure
P
type
(PT),
particle
size
distributions,
fall
velocity
(V).
The
liquid
water
content
(LWC)
mean
mass-weighted
diameter
(Dm)
were
derived
data
during
ZP
events.
Additionally,
a
Micro
Rain
Radar
(MRR)
an
Pluvio2
gauge
used
part
of
Precipitation
Type
Research
Multi-Scale
Experiment
(WINTRE-MIX)
field
campaign
at
Sorel,
Quebec.
dataset
included
manual
measurements
equivalent
(SWE),
PT,
radiosonde
profiles.
analysis
revealed
that
generally
agreed
in
detecting
However,
tended
overestimate
ZR
underestimate
IPs,
while
showed
superior
detection
R,
ZR,
S.
Conversely,
performed
better
identifying
ZL.
These
discrepancies
may
stem
from
uncertainties
velocity–diameter
(V-D)
relationship
diagnose
IPs.
Observations
MRR,
radiosondes,
surface
linked
IP
events
melting
layers
(MLs).
associated
colder
temperatures
(Ts)
compared
Most
ZL
occurrences
characterized
by
light
low
LWC
specific
intensity
Dm
thresholds.
more
common
warmer
T
under
relative
humidity
conditions.
significantly
underestimated
snowfall
optical
probes
measurements.
estimates
data,
adjusted
density
account
riming
effects,
closely
matched
observations.
VFAST Transactions on Software Engineering,
Journal Year:
2025,
Volume and Issue:
13(1), P. 72 - 87
Published: March 4, 2025
Rainfall,
is
one
of
the
most
important
meteorological
factors
that
affects
many
parts
our
everyday
lives
including
crop
productivity,
water
quality,
livestock
availability,
hydroelectric
power
generation
to
name
a
few.
Rainfall
prediction
can
significantly
contribute
boosting
economy
by
enabling
better
planning,
risk
management,
and
resource
allocation
in
various
industrial
sectors.
In
this
study,
forty
years
monsoon
precipitation
data
gathered
for
39
stations
across
five
zones
Pakistan.
We
propose
multi-step
Long
Short-Term
Memory
(LSTM)-based
model
capable
forecasting
Monsoon
yearly
data.
Three
LSTM
models
stack,
bidirectional
convolutional
are
applied
on
dataset
performance
these
analysed
using
centralized
decentralized
approach.
It
observed
RMSE
score
strategy
was
found
than
approach,
whereby
100%
had
lower
as
compared
one.
Moreover,
approach
78.7%
different
exhibited
R2
>
0.9
values
indicating
general
fit
model.