I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
15
, N
o.
1
,
F
e
br
ua
r
y
2026
, pp.
559
~
567
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
15
.i
1
.pp
559
-
567
559
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
E
val
u
at
i
on
of
ar
t
i
f
i
c
i
al
i
n
t
e
l
l
i
ge
n
c
e
al
gor
i
t
h
m
s t
o e
st
i
m
at
e
w
at
e
r
q
u
al
i
t
y p
ar
a
m
e
t
e
r
s u
si
n
g sat
e
l
l
i
t
e
i
m
age
s
Ju
li
o C
e
s
ar
A
n
aya
-
V
al
e
n
z
u
e
la
, G
lo
r
ia
Y
an
e
t
h
F
lo
r
e
z
-
Y
e
p
e
s
, Y
e
is
on
A
lb
e
r
t
o G
ar
c
é
s
-
G
ó
m
e
z
F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng a
nd A
r
c
hi
t
e
c
t
ur
e
, U
ni
ve
r
s
i
da
d C
a
t
ól
i
c
a
de
M
a
ni
z
a
l
e
s
,
M
a
ni
z
a
l
e
s
, C
ol
om
bi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
A
ug
13, 2024
R
e
vi
s
e
d
D
e
c
18, 2025
A
c
c
e
pt
e
d
J
a
n 10, 2026
The
Ciénaga
de
la
Virgen
(Virge
n
Swamp)
is
a
coastal
lagoon
in
Car
tagena
de
Indias
that
provid
es
multiple
e
cosystem
services
in
northern
Bolíva
r.
This
ecosystem
has
faced
anthropog
enic
pressure
from
city
growth
and
im
proper
water
resource
management,
including
wastewater
and
agroch
emical
discharges.
Consequently,
environmental
authorities
must
monitor
certain
sites within the water
body and
extrapolate the
data
across its
entire ex
panse.
In
this
study,
predictive
tools
are
applied
to
determine
water
quality
parameters
such
as
chlorophy
ll
-
a
(CL
-
a)
,
dissolved
oxygen
(DO)
,
total
suspended
solids
(TSS)
,
and
salinity
.
This
is
achieved
by
corr
elating
traditionally
obtained
data
with
the
spectral
r
esponse
of
medium
-
res
olution
satellite
images,
adjusted
using
artificial
intell
igence
(AI)
algor
ithms
.
S
upport
vector
machine
(SVM)
algorit
hms were
used for
regressio
n, r
andom
forests
(RF)
,
and
artificial
neural
networks
(ANN)
,
achieving
an
accuracy
of
79%
for
CL
-
a,
95%
for
DO
,
89%
for
TSS
,
and
96%
for
salinit
y.
Vali
dation
was
performed
using
mean
absolute
percentage
error
(MAPE)
statistical
metrics
and root
mean squ
are error (
RMSE
)
.
K
e
y
w
o
r
d
s
:
D
e
e
p l
e
a
r
ni
ng
M
a
c
hi
ne
l
e
a
r
ni
ng
R
e
m
ot
e
s
e
n
s
in
g
S
pe
c
tr
a
l
s
ig
na
tu
r
e
W
a
te
r
qua
li
ty
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
Y
e
is
on A
lb
e
r
to
G
a
r
c
é
s
-
G
óm
e
z
M
a
s
te
r
i
n R
e
m
ot
e
S
e
n
s
in
g,
F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g a
nd A
r
c
hi
te
c
tu
r
e
, U
ni
ve
r
s
id
a
d C
a
tó
li
c
a
d
e
M
a
ni
z
a
le
s
C
r
a
23 No 60
-
63, M
a
ni
z
a
le
s
, C
a
ld
a
s
, C
ol
om
bi
a
E
m
a
il
:
yga
r
c
e
s
@
uc
m
.e
du.c
o
1.
I
N
T
R
O
D
U
C
T
I
O
N
R
e
m
ot
e
s
e
n
s
in
g
ha
s
pr
ove
n
to
b
e
a
n
im
por
ta
nt
te
c
hnol
ogi
c
a
l
t
ool
in
c
a
pt
ur
in
g
in
f
or
m
a
ti
on
f
r
om
th
e
e
a
r
th
'
s
s
ur
f
a
c
e
[
1]
,
a
id
in
g
in
th
e
m
a
na
g
e
m
e
nt
a
nd
s
us
ta
in
a
b
le
us
e
of
n
a
tu
r
a
l
r
e
s
our
c
e
s
.
O
pt
ic
a
l
s
a
te
ll
it
e
im
a
ge
s
c
ont
r
ib
ut
e
to
th
e
m
oni
to
r
in
g
of
w
a
te
r
r
e
s
our
c
e
s
[
2]
,
pr
ovi
di
ng
a
s
uppor
t
in
s
tr
um
e
nt
th
a
t,
a
lo
ng
w
it
h
ti
m
e
ly
c
a
pt
ur
e
d
da
ta
,
a
ll
ow
s
th
e
m
ode
li
ng
of
th
e
be
ha
vi
or
of
w
a
te
r
qua
li
ty
va
r
ia
bl
e
s
.
I
n
th
is
a
r
ti
c
le
,
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
a
lg
or
i
th
m
s
a
r
e
a
ppl
ie
d
a
nd
a
dj
us
te
d
w
it
h
f
i
e
ld
m
e
a
s
ur
e
m
e
nt
s
[
3]
to
c
a
lc
ul
a
te
th
e
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
s
of
c
hl
or
ophy
ll
-
a
(
C
L
-
a)
,
di
s
s
ol
ve
d
oxyge
n
(
D
O
)
,
to
ta
l
s
us
pe
nde
d
s
ol
id
s
(
T
S
S
)
,
a
nd
s
a
li
ni
ty
.
T
he
s
tu
dy
ut
il
iz
e
s
s
a
te
ll
it
e
im
a
ge
s
of
m
e
di
um
s
pa
ti
a
l
r
e
s
ol
ut
io
n
f
r
om
th
e
C
ié
na
ga
de
la
V
ir
ge
n
(
V
ir
ge
n
S
w
a
m
p)
,
lo
c
a
te
d
in
th
e
c
it
y
o
f
C
a
r
ta
ge
na
de
I
ndi
a
s
in
nor
th
e
r
n
B
ol
ív
a
r
.
T
hi
s
body
of
w
a
te
r
s
us
ta
in
s
m
a
ny
f
a
m
il
ie
s
li
vi
ng
in
it
s
a
r
e
a
of
in
f
lu
e
nc
e
[
4]
,
w
ho
a
r
e
a
f
f
e
c
te
d
by
th
e
de
te
r
io
r
a
ti
on
of
th
e
w
a
te
r
s
due
to
a
nt
hr
opi
c
pr
e
s
s
ur
e
[
5]
,
m
a
in
ly
c
a
us
e
d
by
in
a
de
qua
t
e
s
e
w
a
g
e
di
s
c
ha
r
ge
s
a
nd
s
ol
id
w
a
s
te
,
a
m
ong
ot
he
r
pr
obl
e
m
s
of
th
is
le
nt
ic
body.
A
c
om
pa
r
i
s
on
w
il
l
be
m
a
d
e
i
n
th
e
pe
r
f
or
m
a
n
c
e
of
th
e
pr
op
os
e
d
AI
a
lg
or
it
hm
s
,
in
c
lu
di
ng
s
upp
or
t
ve
c
to
r
m
a
c
hi
ne
s
(
S
V
M
)
f
or
r
e
gr
e
s
s
io
n
,
r
a
ndom
f
or
e
s
t
s
(
R
F
)
(
c
l
a
s
s
if
ie
d
un
de
r
m
a
c
h
in
e
le
a
r
ni
n
g
(
M
L
)
m
e
th
od
s
)
,
a
nd
a
r
ti
f
ic
i
a
l
ne
ur
a
l
n
e
twor
k
s
(
A
N
N
)
(
pa
r
t
of
d
e
e
p
le
a
r
n
in
g
(
D
L
)
)
[
6]
.
T
hi
s
s
tu
dy
f
oc
us
e
s
o
n
c
a
l
c
ul
a
ti
ng
th
e
c
ont
in
uou
s
va
r
i
a
bl
e
s
of
CL
-
a
,
T
S
S
,
a
nd
s
a
li
ni
ty
,
c
ho
s
e
n
f
or
th
e
ir
opt
i
c
a
l
pr
o
pe
r
ti
e
s
[
7]
.
DO
is
a
ls
o
in
c
lu
de
d
,
s
e
le
c
te
d
f
or
it
s
c
r
it
i
c
a
l
r
ol
e
a
s
a
n
in
di
c
a
to
r
of
t
he
a
qua
ti
c
e
c
os
ys
t
e
m
'
s
c
a
p
a
c
i
ty
to
s
uppor
t
f
lo
r
a
a
nd
f
a
u
na
[
6]
.
T
he
e
s
ti
m
a
ti
on
of
th
e
s
e
p
a
r
a
m
e
te
r
s
th
r
ough
a
lg
or
it
hm
s
s
e
r
v
e
s
to
c
om
pl
e
m
e
nt
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
15
, N
o.
1
,
F
e
br
ua
r
y
20
26
:
559
-
567
560
m
a
na
g
e
m
e
nt
of
w
a
t
e
r
r
e
s
our
c
e
s
,
w
it
h
out
in
t
e
ndi
n
g
to
r
e
pl
a
c
e
tr
a
di
ti
o
na
l
m
e
th
od
ol
ogi
e
s
th
a
t
a
r
e
gl
ob
a
ll
y
s
ta
n
da
r
di
z
e
d
a
nd
c
e
r
ti
f
i
e
d
[
8]
,
c
on
duc
t
e
d b
y
s
pe
c
ia
li
z
e
d
l
a
b
or
a
to
r
ie
s
i
n w
a
te
r
q
ua
li
ty
m
e
a
s
ur
e
m
e
n
t.
F
ur
th
e
r
m
or
e
,
r
e
c
e
nt
a
dva
nc
e
m
e
nt
s
in
ML
of
f
e
r
ne
w
a
ve
nue
s
f
or
e
nha
nc
in
g
m
ode
l
r
obus
tn
e
s
s
,
pa
r
ti
c
ul
a
r
ly
w
he
n
de
a
li
ng
w
it
h
li
m
it
e
d
or
no
is
y
da
ta
s
e
ts
.
M
e
th
odol
ogi
e
s
s
uc
h
a
s
s
e
lf
-
s
upe
r
vi
s
e
d
le
a
r
ni
ng
(
S
S
L
)
s
how
pr
om
is
e
in
im
pr
ovi
ng
pe
r
f
or
m
a
nc
e
by
le
ve
r
a
gi
ng
th
e
la
r
ge
a
m
ount
s
of
unl
a
be
le
d
s
a
te
ll
it
e
da
ta
a
va
il
a
bl
e
.
A
ddi
ti
ona
ll
y,
th
e
in
te
gr
a
ti
on
of
da
ta
th
r
ough
m
ul
ti
-
s
e
ns
or
f
us
io
n
te
c
hni
que
s
m
a
y
pr
ovi
de
a
m
or
e
c
om
pr
e
he
ns
iv
e
s
pe
c
tr
a
l
unde
r
s
ta
ndi
ng, pote
nt
ia
ll
y i
m
pr
ovi
ng t
he
a
c
c
ur
a
c
y of
w
a
te
r
qua
li
ty
pr
e
di
c
ti
ons
ove
r
a
r
e
li
a
nc
e
on
a
s
in
gl
e
da
ta
s
our
c
e
.
T
hi
s
s
tu
dy
a
im
s
to
e
va
lu
a
te
t
he
pe
r
f
or
m
a
nc
e
of
va
r
io
us
A
I
a
lg
o
r
it
hm
s
.
T
he
e
va
lu
a
ti
on
f
oc
us
e
s
on
e
s
ti
m
a
ti
ng C
L
-
a
,
D
O
,
T
S
S
,
a
nd
s
a
li
ni
ty
p
a
r
a
m
e
te
r
s
a
s
in
di
c
a
to
r
s
of
w
a
te
r
qua
li
ty
in
th
e
V
ir
ge
n
S
w
a
m
p
.
T
hi
s
e
v
a
lu
a
ti
on
w
il
l
ut
il
iz
e
m
e
di
um
s
p
a
ti
a
l
r
e
s
ol
ut
io
n
s
a
te
ll
it
e
im
a
ge
s
obt
a
in
e
d
f
r
om
th
e
G
oogl
e
E
a
r
th
E
ngi
ne
a
nd J
upyt
e
r
N
ot
e
book pla
tf
or
m
.
2.
M
E
T
H
O
D
2.1. E
s
t
ab
li
s
h
m
e
n
t
of
t
h
e
ar
e
a o
f
i
n
t
e
r
e
s
t
an
d
s
am
p
li
n
g p
oi
n
t
s
T
he
V
ir
ge
n
S
w
a
m
p
is
s
it
ua
t
e
d
in
th
e
c
it
y
of
C
a
r
ta
ge
na
de
I
ndi
a
s
,
in
th
e
D
e
pa
r
tm
e
nt
of
B
ol
ív
a
r
,
c
ove
r
in
g
a
to
ta
l
a
r
e
a
of
502.45
km
².
I
t
is
a
c
oa
s
ta
l
la
goon
s
e
p
a
r
a
te
d
f
r
om
th
e
s
e
a
by
a
s
a
nd
b
a
r
r
ie
r
be
twe
e
n
400 a
nd 800 mete
r
s
w
id
e
t
ha
t
s
ta
r
ts
i
n t
he
vi
ll
a
ge
of
L
a
B
oqui
ll
a
. I
t
ha
s
a
t
r
ia
ngul
a
r
s
ha
pe
, w
it
h a
w
id
th
t
o t
he
s
out
h
of
4.5
km
a
nd
a
le
ngt
h
of
7
km
.
T
he
lo
c
a
ti
on
a
nd
ge
n
e
r
a
l
c
ha
r
a
c
te
r
is
ti
c
s
of
th
e
s
tu
dy
a
r
e
a
a
r
e
pr
e
s
e
nt
e
d
in
F
ig
ur
e
1.
I
n
pa
r
ti
c
ul
a
r
,
t
he
body
of
w
a
te
r
of
th
e
C
ié
na
ga
de
la
V
ir
ge
n
m
e
a
s
ur
e
s
a
ppr
oxi
m
a
te
ly
22.5
km
2
a
nd
ha
s
de
pt
hs
of
up
to
1.6
m
[
9
]
a
s
s
how
n
in
F
ig
u
r
e
1
(
a
)
.
I
n
o
r
de
r
to
m
a
ke
us
e
of
th
e
in
f
o
r
m
a
ti
on
pr
ov
id
e
d,
it
w
a
s
ne
c
e
s
s
a
r
y
to
ge
or
e
f
e
r
e
nc
e
th
e
lo
c
a
ti
on
pl
a
n
pr
ovi
de
d,
s
i
nc
e
th
e
da
ta
di
d
not
ha
ve
th
e
e
xa
c
t
c
oor
di
na
te
of
th
e
s
a
m
pl
in
g
in
th
e
f
ie
ld
.
T
he
ge
or
e
f
e
r
e
nc
in
g
of
th
e
pl
a
ne
w
a
s
c
onduc
t
e
d
to
c
lo
s
e
ly
m
a
tc
h
th
e
ge
om
e
tr
y
of
th
e
body
of
w
a
te
r
us
in
g
id
e
nt
if
ia
bl
e
s
in
uos
it
y
on
th
e
s
hor
e
li
ne
.
T
he
pl
a
ne
'
s
gr
id
us
e
s
a
r
bi
tr
a
r
y
c
oor
di
na
te
s
,
w
hi
c
h
pos
e
d
c
ha
ll
e
ng
e
s
in
a
c
hi
e
vi
ng
pr
e
c
is
e
g
e
or
e
f
e
r
e
nc
in
g.
T
hi
s
li
m
it
a
ti
on
a
f
f
e
c
te
d
th
e
pos
it
io
na
l
a
c
c
ur
a
c
y
of
th
e
s
a
m
pl
in
g
poi
nt
s
,
le
a
di
ng
to
unc
e
r
ta
in
ti
e
s
in
th
e
p
r
e
di
c
ti
on
m
ode
ls
.
A
f
te
r
ge
or
e
f
e
r
e
nc
in
g
th
e
pl
a
n,
th
e
la
bor
a
to
r
y
di
gi
ti
z
e
d
10
s
a
m
pl
in
g
s
it
e
s
w
it
hi
n
th
e
body
of
w
a
te
r
id
e
nt
if
ie
d
by
num
be
r
s
:
2,
4,
5,
6,
7,
8,
10,
22,
28,
a
nd
32.
F
ig
ur
e
1
(
b
)
s
how
s
th
e
lo
c
a
ti
on
pl
a
n
pr
ovi
de
d
by
th
e
C
A
R
D
I
Q
U
E
la
bor
a
to
r
y.
T
o
e
ns
ur
e
th
e
e
xpe
r
im
e
nt
a
l
s
e
tu
p
is
c
le
a
r
f
or
r
e
pl
ic
a
ti
on,
th
e
ge
or
e
f
e
r
e
nc
in
g
pr
oc
e
s
s
,
de
s
pi
te
it
s
c
ha
ll
e
nge
s
,
w
a
s
c
onduc
te
d
a
s
f
ol
lo
w
s
.
T
he
s
c
a
nn
e
d
la
bor
a
to
r
y
pl
a
n
,
a
s
s
how
n
in
F
i
gur
e
1
(
b
)
,
w
a
s
im
por
te
d
in
to
a
ge
ogr
a
phi
c
in
f
or
m
a
ti
on
s
ys
te
m
(
G
I
S
)
e
nvi
r
onm
e
nt
.
I
de
nt
i
f
ia
bl
e
s
hor
e
li
ne
f
e
a
tu
r
e
s
a
nd
s
in
uo
s
it
y
vi
s
ib
le
in
bot
h
th
e
pl
a
n
a
nd
ba
s
e
li
ne
s
a
te
ll
it
e
im
a
ge
r
y
w
e
r
e
u
s
e
d
a
s
gr
ound
c
ont
r
ol
poi
nt
s
to
a
li
gn
th
e
pl
a
n'
s
a
r
bi
tr
a
r
y
gr
id
to
th
e
r
e
a
l
-
w
or
ld
c
oor
di
na
te
s
ys
te
m
.
A
lt
hough
th
is
m
a
nua
l
a
li
gnm
e
nt
in
tr
oduc
e
s
pos
it
io
na
l
unc
e
r
ta
in
ty
,
th
e
10
di
gi
ti
z
e
d s
a
m
pl
in
g point
s
r
e
pr
e
s
e
nt
t
he
be
s
t
a
va
il
a
bl
e
a
ppr
oxi
m
a
ti
on of
t
he
i
n
-
s
it
u c
ol
le
c
ti
on s
it
e
s
.
(
a
)
(
b)
F
ig
ur
e
1. L
oc
a
ti
on of
t
he
s
tu
dy a
r
e
a
of
(
a
)
V
ir
ge
n S
w
a
m
p
a
nd
(
b)
s
a
m
pl
in
g point
s
i
n t
he
s
w
a
m
p
2.2. Ve
r
if
ic
at
io
n
an
d
ad
j
u
s
t
m
e
n
t
of
e
xi
s
t
in
g i
n
f
or
m
at
io
n
of
t
h
e
ar
e
a of
i
n
t
e
r
e
s
t
T
he
r
e
gi
ona
l
a
ut
onomou
s
c
or
por
a
ti
on
of
th
e
C
a
na
l
de
l
D
iq
ue
-
C
a
r
di
que
pr
ovi
de
d
th
e
r
e
s
ul
ts
of
w
a
t
e
r
qua
li
ty
a
na
ly
s
is
f
or
th
e
V
ir
ge
n
S
w
a
m
p
f
r
om
2015
to
2021.
T
h
e
da
ta
c
om
pr
is
e
a
to
ta
l
of
37
s
a
m
pl
in
g
r
e
c
or
ds
f
r
om
bot
h
th
e
body
of
w
a
te
r
a
nd
th
e
c
ha
nne
ls
th
a
t
f
e
e
d
in
to
th
e
s
w
a
m
p.
T
h
e
in
f
or
m
a
ti
on
ha
s
a
n
id
e
nt
if
ie
r
pe
r
poi
nt
t
ha
t
is
l
is
te
d on a
dr
a
w
in
g pr
ovi
de
d by the
C
A
R
D
I
Q
U
E
l
a
bor
a
to
r
y.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
v
al
uat
io
n of
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
al
gor
it
hm
s
t
o e
s
ti
m
at
e
w
at
e
r
quali
ty
…
(
J
ul
io
C
e
s
a
r
A
nay
a
-
V
al
e
nz
ue
la
)
561
2.3. S
e
le
c
t
io
n
of
s
at
e
ll
it
e
i
m
age
s
T
he
da
t
e
s
of
th
e
f
il
te
r
e
d
im
a
ge
s
m
a
tc
h
th
e
d
a
te
s
of
f
ie
ld
s
a
m
pl
e
c
ol
le
c
ti
on
by
th
e
la
bor
a
to
r
y.
T
hi
s
a
ppr
oa
c
h
e
ns
ur
e
d
th
a
t
th
e
s
pe
c
tr
a
l
r
e
s
pon
s
e
of
th
e
s
it
e
s
of
in
te
r
e
s
t
c
or
r
e
s
ponde
d
c
lo
s
e
ly
to
th
e
w
a
te
r
c
ondi
ti
ons
a
na
ly
z
e
d
by
th
e
la
bor
a
to
r
y
[
10]
,
th
e
r
e
by
m
in
im
iz
in
g
unc
e
r
ta
in
ty
r
e
la
te
d
to
th
e
dyna
m
ic
s
of
th
e
s
w
a
m
p'
s
w
a
te
r
body. I
t
is
c
r
uc
ia
l
t
o c
ons
id
e
r
t
he
t
im
e
ga
p be
tw
e
e
n i
m
a
ge
c
a
pt
ur
e
a
nd l
a
bor
a
to
r
y s
a
m
pl
in
g, a
s
w
a
te
r
c
ondi
ti
ons
a
r
e
s
ubj
e
c
t
to
c
ons
ta
nt
c
ha
nge
s
in
f
lu
e
nc
e
d
by
f
a
c
to
r
s
s
uc
h
a
s
di
s
c
ha
r
ge
s
,
te
m
pe
r
a
tu
r
e
va
r
ia
ti
ons
, une
xpe
c
te
d r
a
in
f
a
ll
, a
nd othe
r
a
nt
hr
opoge
ni
c
or
e
nvi
r
onm
e
nt
a
l
c
ondi
ti
ons
i
n t
he
a
r
e
a
[
11]
.
O
ut
of
th
e
37
f
ie
ld
s
a
m
pl
in
g
da
te
s
,
7
im
a
ge
s
or
im
a
ge
c
ol
le
c
ti
ons
w
e
r
e
f
ound
to
ha
ve
c
oi
nc
id
in
g
or
c
lo
s
e
ly
m
a
tc
hi
ng
c
a
pt
ur
e
da
te
s
.
T
he
s
e
im
a
ge
s
or
im
a
ge
c
ol
le
c
ti
ons
ha
ve
a
s
pa
ti
a
l
r
e
s
ol
ut
io
n
of
3
m
e
te
r
s
,
a
r
a
di
om
e
tr
ic
r
e
s
ol
ut
io
n
of
16
bi
ts
,
a
r
e
or
th
or
e
c
ti
f
ie
d,
a
nd
c
or
r
e
c
te
d
to
s
ur
f
a
c
e
r
e
f
le
c
ta
nc
e
[
12]
,
f
a
c
il
it
a
ti
ng
di
r
e
c
t
a
na
ly
s
is
.
A
to
ta
l
of
s
e
ve
n
im
a
ge
c
a
pt
ur
e
da
te
s
w
e
r
e
id
e
nt
if
ie
d
th
a
t
c
lo
s
e
ly
a
li
gne
d
w
it
h
th
e
la
bor
a
to
r
y
s
a
m
pl
in
g
da
te
s
,
e
ns
ur
in
g
te
m
por
a
l
c
on
s
is
te
nc
y
be
twe
e
n
th
e
r
e
m
ot
e
s
e
ns
in
g
d
a
ta
a
nd
th
e
f
ie
ld
m
e
a
s
ur
e
m
e
nt
s
.
T
hi
s
c
or
r
e
s
ponde
nc
e
i
s
c
r
it
ic
a
l
f
or
m
in
im
iz
in
g unc
e
r
ta
in
ty
r
e
la
te
d t
o t
he
dyna
m
ic
na
tu
r
e
of
t
he
s
w
a
m
p'
s
w
a
te
r
body.
T
he
s
pe
c
if
ic
pa
ir
in
gs
of
s
a
te
ll
it
e
im
a
ge
da
te
s
a
nd
la
bor
a
t
or
y
c
ol
le
c
ti
on
da
te
s
ut
il
iz
e
d
f
or
th
is
s
tu
dy
a
r
e
de
ta
il
e
d i
n T
a
bl
e
1.
T
a
bl
e
1. D
a
te
of
c
a
pt
ur
e
of
t
he
s
a
te
ll
it
e
i
m
a
ge
ve
r
s
u
s
da
te
of
c
ol
le
c
ti
on of
t
he
s
a
m
pl
e
by t
he
l
a
bor
a
to
r
y
S
a
t
e
l
l
i
t
e
i
m
a
ge
da
t
e
D
a
t
e
of
s
a
m
pl
e
c
ol
l
e
c
t
i
on by t
he
l
a
bor
a
t
or
y
06/
20/
2017
06/
21/
2017
07/
22/
2017
07/
27/
2017
09/
2/
2017
08/
31/
2017
09/
11/
2017
09/
28/
2017
07/
22/
2018
07/
23/
2018
08/
21/
2018
08/
21/
2018
03/
26/
2019
03/
27/
2019
2.4. P
r
e
p
ar
in
g an
d
u
p
lo
a
d
in
g i
n
f
or
m
at
io
n
t
o t
h
e
G
oogl
e
E
a
r
t
h
E
n
gi
n
e
p
la
t
f
or
m
T
he
G
oogl
e
E
a
r
th
E
ngi
ne
is
a
c
lo
ud
-
ba
s
e
d
pl
a
tf
or
m
de
s
ig
ne
d
f
or
ge
os
pa
ti
a
l
da
ta
a
n
a
ly
s
is
,
w
id
e
ly
ut
il
iz
e
d
by
r
e
s
e
a
r
c
he
r
s
gl
oba
ll
y
f
or
tr
e
nd
a
n
a
ly
s
is
[
13]
.
I
t
le
ve
r
a
ge
s
G
oogl
e
'
s
r
obus
t
c
om
put
in
g
in
f
r
a
s
tr
uc
tu
r
e
to
s
uppor
t
va
r
io
us
r
e
s
e
a
r
c
h
dom
a
in
s
[
14]
. T
o
de
te
r
m
in
e
th
e
a
r
e
a
of
in
te
r
e
s
t,
th
e
in
it
ia
l
s
t
e
p
in
vol
ve
d
u
s
in
g
th
e
ge
om
e
tr
y
in
s
ha
pe
f
il
e
f
or
m
a
t,
s
pe
c
if
ic
a
ll
y
th
e
de
ta
il
e
d
pe
r
m
a
n
e
nt
c
ha
nne
l
of
th
e
V
ir
ge
n
S
w
a
m
p
.
T
hi
s
da
ta
w
a
s
de
r
iv
e
d
f
r
om
te
c
hni
c
a
l
s
tu
di
e
s
c
onduc
te
d
by
C
A
R
D
I
Q
U
E
in
2021,
w
hi
c
h
de
li
ne
a
te
d
th
e
w
a
te
r
pe
r
im
e
te
r
of
th
e
s
w
a
m
p
a
nd
in
te
r
na
l
w
a
te
r
bodi
e
s
w
it
hi
n
C
a
r
ta
ge
na
.
T
h
e
ge
om
e
tr
y
w
a
s
upl
oa
de
d
to
th
e
G
oogl
e
E
a
r
th
E
ngi
ne
pl
a
tf
or
m
a
s
a
s
s
e
ts
a
lo
ng
w
it
h
th
e
poi
nt
ge
om
e
tr
y
w
it
h
th
e
la
bor
a
to
r
y
in
f
or
m
a
ti
on
in
s
ha
pe
f
il
e
f
or
m
a
t
a
nd
th
e
s
e
le
c
te
d
im
a
g
e
c
ol
le
c
ti
ons
w
it
h
ta
ke
da
t
e
06/
20/
2017,
07/
22/
2017,
09/
2/
2017,
09/
11/
2017,
07/
22/
2018, 08/21/
2018
,
a
nd 03/26/
2019.
2.5. Ob
t
ai
n
in
g
n
u
m
e
r
ic
al
m
od
e
ls
T
o
de
te
r
m
in
e
th
e
a
lg
or
it
hm
s
c
or
r
e
la
ti
ng
CL
-
a
,
DO
,
T
S
S
,
a
n
d
s
a
li
ni
ty
c
on
c
e
nt
r
a
ti
ons
,
th
e
s
tu
dy
s
e
le
c
te
d
f
our
im
a
ge
ba
nd
s
:
b1
(
bl
ue
,
0.455
-
0.515
µm
)
,
b2
(
gr
e
e
n
,
0.5
-
0.59
µm
)
,
b3
(
r
e
d
,
0.59
-
0.67
µm
)
,
a
n
d
b4
(
ne
a
r
-
in
f
r
a
r
e
d
(
N
I
R
)
,
0.78
-
0.86
µm
)
.
T
he
s
e
ba
nds
a
r
e
c
om
m
onl
y
us
e
d
in
w
a
te
r
body
s
tu
di
e
s
[
2]
a
nd
w
e
r
e
c
hos
e
n
a
s
in
d
e
pe
nde
nt
or
r
e
gr
e
s
s
io
n
v
a
r
ia
bl
e
s
.
A
c
c
or
di
ng
to
B
r
ic
e
ño
e
t
al
.
[
15]
,
a
s
w
a
ve
le
ngt
h
in
c
r
e
a
s
e
s
,
w
a
te
r
a
bs
or
bs
m
or
e
in
c
id
e
nt
e
n
e
r
gy,
r
e
s
ul
ti
ng
in
lo
w
e
r
or
n
e
gl
ig
ib
le
e
ne
r
gy
r
e
f
le
c
ti
on
be
yond
th
e
N
I
R
b
a
nds
,
w
hi
c
h
th
e
r
e
f
or
e
do
no
t
c
ont
r
ib
ut
e
s
ig
ni
f
ic
a
nt
ly
to
w
a
te
r
qua
li
ty
a
na
ly
s
is
[
16]
.
L
ik
e
w
is
e
,
th
e
s
tu
dy
c
ons
id
e
r
e
d
th
e
nor
m
a
li
z
e
d
di
f
f
e
r
e
nc
e
ve
g
e
ta
ti
on
in
de
x
(
N
D
V
I
)
,
th
e
n
or
m
a
li
z
e
d
di
f
f
e
r
e
nc
e
w
a
te
r
in
de
x
(
N
D
W
I
)
,
1
(
0
.
455
−
0
.
515
µ
)
2
(
0
.
5
−
0
.
59
µ
)
,
2
(
0
.
5
−
0
.
59
µ
)
3
(
0
.
59
−
0
.
67
µ
)
,
2
(
0
.
5
−
0
.
59
µ
)
4
(
0
.
78
−
0
.
86
µ
)
,
3
(
0
.
59
−
0
.
67
µ
)
4
(
0
.
78
−
0
.
86
µ
)
a
nd
s
im
pl
e
r
a
ti
os
ba
s
e
d
on
s
p
e
c
tr
a
l
s
ig
na
tu
r
e
a
na
ly
s
i
s
a
t
di
f
f
e
r
e
nt
le
ve
ls
of
in
c
id
e
nt
e
n
e
r
gy,
a
s
c
onduc
te
d
by
R
uddi
c
k
e
t
al
.
[
17]
.
F
or
th
e
c
a
lc
ul
a
ti
on
of
CL
-
a
,
pr
om
in
e
nt
pe
a
k
s
in
th
e
gr
e
e
n
r
e
gi
on
(
b2,
0
.5
-
0.59
μ
m
)
a
nd
e
ne
r
gy
a
b
s
or
pt
i
on
in
t
he
bl
ue
(
b1,
0.45
5
-
0.515
μ
m
)
a
nd
r
e
d
(
b3,
0.59
-
0.
67
μ
m
)
r
e
gi
ons
a
r
e
obs
e
r
ve
d
i
n
th
e
s
p
e
c
tr
a
l
s
ig
na
t
ur
e
.
F
or
T
S
S
,
a
bs
or
pt
io
n
i
s
not
ic
e
a
b
le
in
th
e
bl
ue
b
a
nd
(
b1,
0.4
55
-
0.5
15
μ
m
)
,
in
c
r
e
a
s
e
s
in
th
e
gr
e
e
n
ba
nd
(
b2
,
0.5
-
0.5
9
μ
m
)
,
pe
a
k
s
i
n
th
e
r
e
d
b
a
nd
(
b3,
0.59
-
0.
67
μ
m
)
,
a
nd
de
c
r
e
a
s
e
s
in
th
e
N
I
R
b
a
nd
(
b
4,
0
.78
-
0.
86
μ
m
)
[
1
8]
.
A
ddi
ti
ona
ll
y,
ne
w
ba
nds
w
e
r
e
s
e
le
c
te
d
f
r
om
th
e
pr
in
c
ip
a
l
c
om
pone
nt
s
to
le
ve
r
a
ge
th
e
ir
lo
w
c
or
r
e
la
ti
on.
W
it
h
th
e
ge
om
e
tr
y c
or
r
e
s
ponding t
o t
he
f
ie
ld
s
a
m
pl
in
g s
it
e
s
, pr
e
vi
ous
ly
l
oa
de
d a
s
a
n
a
s
s
e
t
in
G
oogl
e
E
a
r
th
E
ng
in
e
,
r
e
f
le
c
ta
nc
e
va
lu
e
s
w
e
r
e
e
xt
r
a
c
te
d
f
or
e
a
c
h
of
th
e
in
de
pe
nde
nt
v
a
r
ia
bl
e
s
.
T
he
f
ol
lo
w
in
g
ta
bl
e
li
s
ts
a
s
a
m
pl
e
of
th
e
da
ta
obt
a
in
e
d.
2.6.
D
at
a
a
n
al
ys
is
F
or
d
a
t
a
a
n
a
ly
s
i
s
,
AI
te
c
h
ni
q
ue
s
w
e
r
e
e
m
pl
o
ye
d
u
s
i
ng
ML
m
o
de
l
s
w
it
h
r
e
gr
e
s
s
io
n
,
i
nc
lu
d
in
g
S
V
M
,
R
F
,
a
nd
DL
m
od
e
l
s
s
uc
h a
s
A
N
N
.
T
he
s
e
l
e
c
ti
on of
ba
nd
s
,
in
di
c
e
s
, a
n
d s
im
pl
e
quo
ti
e
nt
s
w
a
s
b
a
s
e
d
o
n
l
it
e
r
a
tu
r
e
s
o
ur
c
e
s
s
uc
h
a
s
t
h
e
s
t
udy
b
y
B
r
i
c
e
ño
e
t
al
.
[
15]
.
T
h
e
ir
a
n
a
ly
s
i
s
of
a
bs
or
p
ti
o
n
a
n
d
r
e
f
l
e
c
ti
on
le
ve
l
s
i
n
s
pe
c
tr
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
15
, N
o.
1
,
F
e
br
ua
r
y
20
26
:
559
-
567
562
s
i
gn
a
tu
r
e
s
de
f
in
e
d
th
e
s
e
v
a
r
i
a
b
le
s
a
s
e
xpl
a
n
a
t
or
y
f
a
c
t
or
s
.
S
im
i
la
r
ly
,
b
a
s
e
d
o
n
th
e
r
e
s
e
a
r
c
h
r
e
s
u
lt
s
of
[
19]
,
w
it
h
m
ul
ti
pl
e
r
e
gr
e
s
s
io
n m
ode
ls
w
hos
e
pr
e
di
c
to
r
va
r
ia
bl
e
s
a
r
e
ba
s
e
d on the
vi
s
ib
le
s
pe
c
tr
um
ba
nds
of
th
e
C
B
E
R
S
-
2B
s
a
te
ll
it
e
.
R
e
g
a
r
di
ng
th
e
ba
nds
de
r
iv
e
d
f
r
om
pr
in
c
ip
a
l
c
o
m
pone
nt
s
a
na
ly
s
i
s
(
P
C
A
)
,
th
e
obj
e
c
ti
ve
is
to
de
te
r
m
in
e
w
he
th
e
r
in
c
or
por
a
ti
ng
th
e
va
r
ia
bi
li
ty
c
ont
r
ib
ut
e
d
by
th
e
s
e
ba
nds
opt
im
iz
e
s
m
ode
l
s
,
a
s
de
m
ons
tr
a
te
d
in
pr
e
vi
ous
s
tu
di
e
s
li
ke
th
a
t
of
L
ope
s
e
t
al
.
[
20
]
.
I
t'
s
c
r
uc
ia
l
to
not
e
th
a
t
c
hl
or
ophyll
a
bs
or
bs
m
or
e
e
le
c
tr
om
a
gne
ti
c
e
ne
r
gy
in
th
e
bl
ue
ba
nd,
w
i
th
a
n
in
c
r
e
a
s
e
in
th
e
gr
e
e
n
ba
nd
[
21]
.
T
he
r
e
f
or
e
,
th
e
s
e
ba
nds
a
r
e
e
s
s
e
nt
ia
l
f
or
in
c
lu
s
io
n
in
num
e
r
ic
a
l
m
ode
ls
.
A
ddi
ti
ona
ll
y,
th
e
r
e
d
a
nd
in
f
r
a
r
e
d
ba
nds
a
r
e
s
ig
ni
f
ic
a
nt
f
or
de
te
c
ti
ng
s
us
pe
nde
d
s
ol
id
s
, a
s
di
s
c
us
s
e
d by
[
22]
.
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
3.1. Re
s
u
lt
of
t
h
e
m
od
e
l
w
it
h
s
u
p
p
or
t
ve
c
t
or
m
ac
h
in
e
T
he
f
i
r
s
t
AI
a
lg
or
it
hm
u
ti
li
z
e
d
w
a
s
S
V
M
f
or
r
e
g
r
e
s
s
io
n.
T
o
pe
r
f
o
r
m
th
e
r
e
g
r
e
s
s
io
n
by
m
e
a
ns
o
f
th
is
a
lg
or
it
hm
,
it
is
ne
c
e
s
s
a
r
y
to
ha
ve
a
tr
a
i
ni
n
g
da
ta
s
e
t
a
n
d
a
t
e
s
t
da
ta
s
e
t
[
3
]
,
s
o
i
n
th
is
s
t
udy
3
0
%
of
th
e
to
t
a
l
da
ta
w
a
s
e
s
ta
b
li
s
he
d
f
or
th
e
te
s
t
d
a
ta
s
e
t.
T
hi
s
pe
r
c
e
nt
a
g
e
is
us
e
d
f
o
r
m
od
e
l
pr
e
di
c
t
io
n
a
nd
va
li
da
t
io
n.
T
he
s
c
i
ki
t
-
le
a
r
n
P
yt
hon
l
ib
r
a
r
y
p
r
ov
id
e
s
t
oo
ls
t
ha
t
s
i
m
p
li
f
y
pr
og
r
a
m
m
i
ng
a
nd
m
a
th
e
m
a
ti
c
a
l
c
a
lc
ul
a
ti
ons
f
o
r
m
o
de
ls
,
i
nc
lu
d
in
g
th
e
S
V
M
l
ib
r
a
r
y
,
w
hi
c
h
in
c
lu
de
s
th
e
s
upp
or
t
ve
c
t
o
r
r
e
g
r
e
s
s
io
n
(
S
V
R
)
a
l
go
r
i
th
m
.
T
he
n,
it
is
ne
c
e
s
s
a
r
y
to
de
f
in
e
a
ke
r
ne
l
f
un
c
t
io
n
,
w
h
ic
h
c
a
n
b
e
li
n
e
a
r
or
no
n
-
li
ne
a
r
.
F
o
r
t
hi
s
p
a
r
ti
c
ul
a
r
c
a
s
e
,
th
e
de
f
a
u
lt
ke
r
n
e
l
f
oun
d,
r
a
di
a
l
ba
s
is
f
unc
t
io
n
(
R
B
F
)
,
w
h
i
c
h
c
or
r
e
s
p
on
ds
to
a
G
a
us
s
ia
n
ke
r
ne
l
,
w
a
s
s
e
le
c
t
e
d.
T
he
s
e
pa
r
a
m
e
te
r
s
w
e
r
e
a
pp
li
e
d
t
o
t
he
f
o
ur
w
a
te
r
q
ua
l
it
y
va
r
i
a
bl
e
s
o
f
in
t
e
r
e
s
t.
I
t
is
i
m
po
r
ta
n
t
t
o
m
e
n
ti
o
n
th
a
t
t
he
da
ta
w
e
r
e
p
r
e
v
io
us
ly
s
ta
nda
r
d
iz
e
d,
s
in
c
e
AI
a
lg
or
it
hm
s
c
a
n
ha
ve
e
r
r
o
ne
o
us
pe
r
f
o
r
m
a
nc
e
if
th
e
da
t
a
do
no
t
f
o
ll
o
w
a
m
o
r
e
o
r
l
e
s
s
no
r
m
a
l
or
G
a
us
s
ia
n
di
s
t
r
ib
u
ti
on
,
f
o
r
t
hi
s
th
e
s
c
ik
it
-
l
e
a
r
n
t
oo
l
w
a
s
us
e
d,
S
ta
n
da
r
d
S
c
a
le
r
w
h
ic
h
e
li
m
in
a
te
s
t
he
m
e
a
n
o
f
t
he
d
a
ta
a
nd
s
c
a
le
s
th
e
m
w
it
h
va
r
ia
nc
e
e
q
ua
l
to
1,
by
m
e
a
ns
o
f
t
he
c
a
lc
u
la
t
io
n
=
(
−
)
⁄
,
w
he
r
e
x
i
t
is
t
he
va
lu
e
of
th
e
t
r
a
in
i
ng
da
ta
,
u
it
is
th
e
m
e
a
n
o
f
t
he
tr
a
in
in
g
s
a
m
pl
e
,
s
th
e
s
ta
n
da
r
d
d
e
vi
a
t
io
n
[
23
]
.
A
f
te
r
r
un
ni
n
g
th
e
m
ode
ls
,
it
is
i
m
p
or
ta
n
t
to
r
e
v
e
r
t
t
o
t
he
in
i
ti
a
l
va
lu
e
s
to
ob
ta
in
da
ta
a
c
c
o
r
di
ng
to
th
e
o
r
ig
i
na
l
s
c
a
le
o
f
th
e
va
r
ia
b
le
s
be
in
g
pr
e
d
ic
te
d.
T
h
is
s
te
p
is
a
c
c
om
pl
is
he
d
us
i
ng
th
e
`
i
nve
r
s
e
_t
r
a
ns
f
o
r
m
`
f
un
c
t
io
n
,
w
h
ic
h
is
a
ls
o
i
nc
o
r
p
or
a
te
d
in
to
th
e
S
ta
nda
r
dS
c
a
le
r
a
lg
o
r
it
hm
.
T
o
c
a
lc
u
la
te
t
he
a
c
c
u
r
a
c
y
of
t
he
m
ode
l
w
it
h
S
V
R
,
th
e
m
e
a
n
a
bs
ol
u
te
pe
r
c
e
n
ta
ge
e
r
r
or
(
M
A
P
E
)
w
a
s
us
e
d
A
gu
il
a
r
a
nd
D
ía
z
[
3
]
.
M
A
P
E
is
de
r
iv
e
d
f
r
om
th
e
m
e
a
n
o
f
th
e
a
bs
o
lu
te
pe
r
c
e
n
ta
g
e
e
r
r
o
r
(
A
P
E
)
,
w
hi
c
h
is
c
a
lc
u
la
te
d
us
i
ng
(
1)
.
=
100
∗
(
(
−
′
)
)
(
1)
W
he
r
e
is
th
e
ob
s
e
r
ve
d
v
a
lu
e
a
nd
′
pr
e
di
c
ti
on
va
lu
e
.
W
it
h
th
e
M
A
P
E
,
th
e
pr
e
di
c
ti
on
of
th
e
m
ode
l
w
a
s
c
a
lc
ul
a
te
d by me
a
n
s
of
(
2)
.
ó
=
100
−
(
2)
T
he
m
ode
l
w
it
h
S
V
R
a
ll
ow
e
d
obt
a
in
in
g
th
e
f
ol
lo
w
in
g
pe
r
f
or
m
a
nc
e
f
or
e
a
c
h
w
a
te
r
qu
a
li
ty
pa
r
a
m
e
te
r
of
in
te
r
e
s
t
(
s
e
e
T
a
bl
e
2)
.
I
t
is
im
por
ta
nt
to
m
e
nt
io
n
th
a
t
th
e
ba
s
is
f
or
th
e
s
e
le
c
ti
on
of
pr
e
di
c
to
r
va
r
ia
bl
e
s
w
a
s
th
e
s
ig
ni
f
ic
a
nt
c
or
r
e
la
ti
on
be
twe
e
n
th
e
s
e
a
nd
th
e
r
e
s
pons
e
va
r
ia
bl
e
.
H
ow
e
ve
r
,
f
r
om
th
is
ba
s
is
,
w
e
pr
oc
e
e
de
d
w
it
h
th
e
e
nt
r
y
of
ot
he
r
va
r
ia
bl
e
s
e
m
pi
r
ic
a
ll
y
or
th
e
e
xc
lu
s
io
n
of
th
e
m
,
s
e
e
ki
ng
th
os
e
th
a
t
im
pr
ove
d
th
e
pe
r
f
or
m
a
nc
e
of
th
e
m
ode
l.
W
hi
le
th
e
in
i
ti
a
l
s
e
le
c
ti
on
of
p
r
e
di
c
to
r
va
r
ia
bl
e
s
w
a
s
ba
s
e
d
on
s
ig
ni
f
ic
a
nt
c
or
r
e
la
ti
ons
id
e
nt
if
ie
d
in
th
e
li
te
r
a
tu
r
e
a
nd
pr
e
li
m
in
a
r
y
a
n
a
ly
s
is
,
th
e
f
in
a
l
va
r
ia
bl
e
s
e
t
s
w
e
r
e
r
e
f
in
e
d
e
m
pi
r
ic
a
ll
y.
T
hi
s
it
e
r
a
ti
ve
a
ppr
oa
c
h
w
a
s
ne
c
e
s
s
a
r
y
to
opt
im
iz
e
m
ode
l
pe
r
f
or
m
a
nc
e
f
or
th
is
s
p
e
c
if
ic
da
ta
s
e
t.
H
ow
e
ve
r
,
w
e
a
c
kno
w
le
dge
th
is
e
m
pi
r
ic
a
l
r
e
f
in
e
m
e
nt
c
a
r
r
ie
s
a
r
is
k
of
ove
r
f
it
ti
ng
a
nd
m
a
y
in
f
lu
e
nc
e
th
e
m
ode
l'
s
ge
ne
r
a
li
z
a
bi
li
ty
.
A
m
or
e
th
e
or
e
ti
c
a
ll
y
gr
ounde
d
f
e
a
tu
r
e
s
e
le
c
ti
on
pr
oc
e
s
s
is
r
e
c
om
m
e
nde
d
f
or
f
ut
ur
e
s
tu
di
e
s
w
it
h l
a
r
ge
r
da
ta
s
e
ts
.
T
a
bl
e
2. M
od
e
l
a
c
c
ur
a
c
y by
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
w
it
h
S
V
M
f
or
r
e
gr
e
s
s
io
n
W
a
t
e
r
qua
l
i
t
y pa
r
a
m
e
t
e
r
P
r
e
di
c
t
or
va
r
i
a
bl
e
s
A
c
c
ur
a
c
y
(%)
M
A
P
E
(%)
R
M
S
E
CL
-
a
'
r
_b1'
, '
r
_b2
'
, '
r
_b3'
,
'
r
_b2_b3'
, '
r
_ndw
i
'
68
32
7.2 μ
g/
L
DO
'
r
_b4'
,'
r
_ndw
i
'
, '
r
_pc
1'
,'
r
_pc
2'
,'
r
_pc
4'
93
7
0.6 m
g O
2/
L
T
S
S
'
r
_b1'
,'
r
_b3
'
, '
r
_b1_b2'
,
'
r
_b3_b4'
, '
r
_b2_b4'
, '
r
_ndvi
'
84
16
18.5 m
g/
L
S
a
l
i
ni
t
y
'
r
_b1'
,'
r
_b2
'
,'
r
_b4'
,
'
r
_b1_b2'
, '
r
_b2_b3'
, '
r
_ndvi
'
, '
r
_pc
1'
94
6
1.9 o/
oo
I
n
or
de
r
to
a
na
ly
z
e
th
e
p
e
r
f
or
m
a
nc
e
of
th
e
m
ode
l
s
w
it
h
S
V
M
f
or
r
e
gr
e
s
s
io
n,
di
s
pe
r
s
io
n
di
a
gr
a
m
s
w
e
r
e
ge
ne
r
a
te
d
be
twe
e
n
th
e
obs
e
r
ve
d
v
a
lu
e
s
a
nd
th
e
pr
e
di
c
te
d
va
lu
e
s
.
T
h
e
s
e
di
a
gr
a
m
s
r
e
por
t
a
gr
oupi
ng
of
th
e
da
ta
w
it
h a
l
in
e
a
r
it
y pa
tt
e
r
n, e
s
pe
c
ia
ll
y i
n t
hos
e
w
hos
e
p
e
r
f
or
m
a
nc
e
e
xc
e
e
d
e
d 90%
, a
s
i
s
t
he
c
a
s
e
w
it
h
DO
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
v
al
uat
io
n of
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
al
gor
it
hm
s
t
o e
s
ti
m
at
e
w
at
e
r
quali
ty
…
(
J
ul
io
C
e
s
a
r
A
nay
a
-
V
al
e
nz
ue
la
)
563
a
nd
s
a
li
ni
ty
(
s
e
e
F
ig
ur
e
2)
.
A
c
c
or
di
ng
to
F
ig
ur
e
2,
a
nd
a
s
e
vi
de
nc
e
d
by
th
e
a
n
a
ly
ti
c
a
l
r
e
s
ul
t
of
th
e
pr
e
di
c
ti
on,
th
e
C
L
-
a
pa
r
a
m
e
te
r
pr
e
s
e
nt
e
d
th
e
lo
w
e
s
t
pe
r
f
or
m
a
nc
e
w
he
n
pr
e
di
c
ti
ng
w
it
h
da
ta
di
f
f
e
r
e
nt
f
r
o
m
th
os
e
us
e
d
in
th
e
t
r
a
in
in
g s
e
t.
H
ow
e
ve
r
, t
he
a
c
c
ur
a
c
y of
68%
a
gr
e
e
s
w
it
h
th
e
r
e
s
ul
t
of
L
e
de
s
m
a
e
t
al
.
[
24]
,
w
hi
c
h t
a
ke
s
i
nt
o
a
c
c
ount
gr
e
e
n a
nd ne
a
r
-
in
f
r
a
r
e
d a
s
r
e
gr
e
s
s
iv
e
va
r
ia
bl
e
s
.
F
ig
ur
e
2. D
is
pe
r
s
io
n
di
a
gr
a
m
s
be
twe
e
n ob
s
e
r
va
ti
on a
nd pr
e
di
c
t
io
n w
it
h
S
V
M
3.2. Re
s
u
lt
of
t
h
e
m
od
e
l
w
it
h
r
an
d
om
f
or
e
s
t
s
T
he
s
e
c
ond
m
od
e
l
us
e
d
is
th
e
RF
.
A
s
in
th
e
S
V
M
,
30%
of
t
he
to
ta
l
s
a
m
pl
e
w
a
s
r
e
s
e
r
ve
d
f
or
th
e
va
li
da
ti
on
of
th
e
m
ode
l.
F
r
o
m
th
e
P
yt
hon
li
b
r
a
r
y
s
c
ik
it
-
le
a
r
n
,
t
he
RF
r
e
gr
e
s
s
or
a
lg
or
it
hm
is
us
e
d,
w
hi
c
h
ha
s
a
s
a
pa
r
a
m
e
te
r
of
s
pe
c
ia
l
im
por
ta
nc
e
th
e
num
b
e
r
of
e
s
ti
m
a
to
r
s
or
tr
e
e
s
th
a
t
m
a
k
e
up
th
e
f
or
e
s
t
to
b
e
bui
lt
.
I
n
a
ddi
ti
on
to
th
e
a
f
or
e
m
e
nt
io
ne
d
pa
r
a
m
e
te
r
,
th
e
RF
r
e
gr
e
s
s
or
li
br
a
r
y
in
c
lu
de
s
a
s
e
r
ie
s
of
ot
he
r
pa
r
a
m
e
te
r
s
,
s
u
c
h
a
s
th
e
c
r
it
e
r
io
n
f
or
s
e
pa
r
a
bi
li
ty
or
th
e
de
pt
h
of
th
e
tr
e
e
s
[
23]
,
a
m
ong
ot
he
r
s
,
w
hi
c
h
w
e
r
e
not
us
e
d
f
or
th
e
pr
e
di
c
ti
on
of
th
e
w
a
te
r
qua
li
ty
in
di
c
a
to
r
s
pr
opos
e
d
in
th
i
s
doc
um
e
nt
.
B
e
f
or
e
ge
ne
r
a
ti
ng
th
e
f
or
e
s
t,
w
e
pr
oc
e
e
de
d
w
it
h
th
e
s
ta
nda
r
di
z
a
ti
on
of
th
e
da
ta
,
a
s
w
a
s
done
w
it
h
th
e
S
V
R
s
.
A
f
te
r
th
e
pr
e
di
c
ti
on,
it
w
a
s
ne
c
e
s
s
a
r
y
to
r
e
s
c
a
le
th
e
da
ta
to
th
e
in
it
ia
l
va
lu
e
s
of
th
e
w
a
te
r
q
ua
li
ty
pa
r
a
m
e
te
r
s
of
in
te
r
e
s
t
in
or
de
r
to
obt
a
in
c
om
pa
r
a
bl
e
in
f
or
m
a
ti
on
a
nd
ve
r
if
y
th
e
a
c
c
ur
a
c
y
of
th
e
m
ode
l.
T
hi
s
a
c
c
ur
a
c
y
w
a
s
c
a
l
c
ul
a
te
d
in
th
e
s
a
m
e
w
a
y
us
in
g
th
e
M
A
P
E
.
M
ode
li
ng w
it
h
RF
pr
e
s
e
nt
s
a
s
ig
ni
f
ic
a
nt
im
pr
ove
m
e
nt
w
it
h
r
e
s
pe
c
t
to
S
V
M
,
w
it
h
a
m
a
r
ke
d
li
ne
a
r
t
r
e
nd be
twe
e
n obs
e
r
va
ti
on a
nd pr
e
di
c
ti
on
a
s
s
how
n i
n
T
a
bl
e
3
.
T
a
bl
e
3. M
od
e
l
a
c
c
ur
a
c
y by
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
w
it
h
RF
f
or
r
e
gr
e
s
s
io
n
W
a
t
e
r
qua
l
i
t
y pa
r
a
m
e
t
e
r
P
r
e
di
c
t
or
va
r
i
a
bl
e
s
A
c
c
ur
a
c
y (
%
)
M
A
P
E
(
%
)
R
M
S
E
CL
-
a
'
r
_b3'
, '
r
_b1_b2
'
79
21
2.7 μ
g/
L
DO
'
r
_b4'
, '
r
_b2_b4
'
,'
r
_pc
4'
95
5
0.5 m
g O
2/
L
T
S
S
'
r
_b1'
,'
r
_b2
'
,'
r
_ndw
i
'
89
11
8.9 m
g/
L
S
a
l
i
ni
t
y
'
r
_b1'
,'
r
_b2_b3
'
,'
r
_b2_b4'
96
4
1.1 o/
oo
3.3.
R
e
s
u
lt
of
t
h
e
m
od
e
l
w
it
h
ar
t
if
ic
ia
l
n
e
u
r
al
n
e
t
w
or
k
s
T
he
la
s
t
m
ode
l
a
ppl
ie
d
c
or
r
e
s
pond
s
to
th
e
ANN
a
nd
in
th
e
s
a
m
e
w
a
y
a
s
in
th
e
two
pr
e
vi
ou
s
m
ode
ls
,
30%
of
th
e
s
a
m
pl
e
da
t
a
w
a
s
r
e
s
e
r
ve
d,
in
or
de
r
to
us
e
th
e
m
a
s
t
e
s
t
or
te
s
t
d
a
ta
f
or
m
ode
l
va
li
da
ti
on.
T
o
w
or
k
w
it
h
DL
in
P
yt
hon,
it
is
e
s
s
e
nt
ia
l
to
ut
il
iz
e
li
br
a
r
ie
s
s
uc
h
a
s
T
h
e
a
no,
T
e
n
s
or
F
lo
w
,
a
nd
K
e
r
a
s
.
T
he
s
e
li
br
a
r
ie
s
a
r
e
de
s
ig
ne
d
f
or
opt
im
iz
in
g
num
e
r
ic
a
l
m
od
e
ls
a
nd
pr
ovi
de
t
ool
s
th
a
t
s
im
pl
if
y
th
e
de
v
e
lo
pm
e
nt
of
ne
ur
a
l
ne
twor
ks
.
I
n
th
e
c
ode
,
w
he
n
K
e
r
a
s
is
im
po
r
te
d
by
de
f
a
ul
t,
T
e
ns
or
f
lo
w
is
lo
a
de
d,
a
nd
w
it
hi
n
K
e
r
a
s
th
e
r
e
a
r
e
two
m
odul
e
s
ne
c
e
s
s
a
r
y
f
or
th
e
c
ons
tr
uc
ti
on
of
th
e
r
e
d
ne
ur
on
a
l
a
r
ti
f
ic
ia
l
(
R
N
A
)
m
ode
l.
T
he
f
ir
s
t
m
odul
e
is
s
e
que
nt
ia
l
,
w
hi
c
h
a
ll
ow
s
you
to
in
it
ia
li
z
e
th
e
n
e
twor
k
pa
r
a
m
e
te
r
s
,
a
nd
th
e
s
e
c
ond
m
odul
e
is
d
e
ns
e
,
w
hi
c
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
15
, N
o.
1
,
F
e
br
ua
r
y
20
26
:
559
-
567
564
a
ll
ow
s
you
to
c
r
e
a
te
th
e
in
te
r
m
e
di
a
t
e
la
ye
r
s
of
th
e
ne
twor
k.
I
n
th
e
de
n
s
e
m
odul
e
,
th
e
uni
ts
pa
r
a
m
e
te
r
w
a
s
ut
il
iz
e
d,
r
e
pr
e
s
e
nt
in
g
th
e
num
be
r
of
node
s
in
th
e
hi
dde
n
la
ye
r
.
T
hi
s
is
a
n
e
m
pi
r
ic
a
l
hype
r
pa
r
a
m
e
te
r
de
te
r
m
in
e
d
by
th
e
d
e
ve
lo
pe
r
.
A
ddi
ti
ona
ll
y,
th
e
in
put
_di
m
pa
r
a
m
e
te
r
,
w
hi
c
h
s
pe
c
if
ie
s
th
e
num
b
e
r
of
in
de
pe
nde
nt
va
r
ia
bl
e
s
,
w
a
s
r
e
pl
a
c
e
d
w
it
h
th
e
e
xpr
e
s
s
io
n
X
_t
r
a
in
.s
ha
pe
[
1]
,
a
c
om
m
a
nd
th
a
t
r
e
tu
r
ns
th
e
to
ta
l
num
be
r
of
e
xpl
a
na
to
r
y
va
r
ia
bl
e
s
.
I
n
th
e
ke
r
ne
l_
in
it
ia
li
z
e
r
,
th
e
f
unc
ti
on
th
a
t
in
it
ia
li
z
e
s
th
e
w
e
ig
ht
s
of
th
e
f
e
a
tu
r
e
a
r
r
a
y
is
e
nt
e
r
e
d
r
a
ndoml
y,
f
or
w
hi
c
h
a
nor
m
a
l
di
s
tr
ib
ut
io
n
f
unc
ti
on
w
a
s
us
e
d
a
nd
th
e
a
c
ti
va
ti
on
f
unc
ti
on
a
ll
ow
s
de
te
r
m
in
in
g
by
m
e
a
ns
of
th
e
w
e
ig
ht
s
w
he
th
e
r
or
not
th
e
in
f
or
m
a
ti
on
f
ol
lo
w
s
th
e
ne
xt
la
ye
r
a
c
c
or
di
ng t
o i
ts
i
m
por
ta
nc
e
.
A
f
te
r
c
onf
ig
ur
in
g
th
e
f
ir
s
t
la
ye
r
,
th
e
s
e
c
ond
hi
dde
n
la
ye
r
w
a
s
a
dde
d
w
it
h
pa
r
a
m
e
te
r
s
s
im
il
a
r
to
th
o
s
e
of
th
e
pr
e
vi
ous
la
ye
r
,
e
x
c
e
pt
f
or
in
put
_di
m
,
w
hi
c
h
w
a
s
a
lr
e
a
d
y
s
pe
c
if
ie
d
in
th
e
f
ir
s
t
la
ye
r
.
I
n
th
is
c
a
s
e
,
th
e
a
c
ti
va
ti
on
f
unc
ti
on
w
a
s
c
ha
nge
d
to
ta
nh.
S
ubs
e
qu
e
nt
ly
,
th
e
out
put
la
ye
r
w
a
s
d
e
f
in
e
d
w
it
h
a
s
in
gl
e
uni
t
a
nd
a
nor
m
a
l
di
s
tr
ib
ut
io
n
in
it
ia
li
z
a
ti
on
ke
r
ne
l.
T
o
c
om
pi
le
th
e
m
od
e
l,
a
lo
s
s
f
unc
ti
on
is
r
e
qui
r
e
d.
T
he
le
a
s
t
s
qu
a
r
e
s
e
r
r
or
f
unc
ti
on
w
a
s
s
e
le
c
te
d
to
m
in
im
iz
e
th
e
di
f
f
e
r
e
nc
e
be
t
w
e
e
n
th
e
obs
e
r
va
ti
ons
a
nd
pr
e
di
c
ti
ons
.
F
or
opt
im
iz
a
ti
on,
th
e
A
da
m
a
lg
or
it
hm
w
a
s
us
e
d.
T
hi
s
opt
im
iz
e
r
he
lp
s
in
f
in
di
ng
th
e
m
os
t
a
c
c
ur
a
te
s
e
t
of
w
e
ig
ht
s
a
nd
is
th
e
de
f
a
ul
t
opt
im
iz
e
r
w
it
hi
n
th
e
c
om
pi
la
ti
on
m
odul
e
.
F
in
a
ll
y,
th
e
m
ode
l
m
us
t
b
e
tr
a
in
e
d.
F
or
th
is
,
you
ne
e
d
th
e
tr
a
in
in
g
da
ta
f
or
th
e
in
de
pe
nde
nt
va
r
ia
bl
e
s
a
nd
th
e
de
pe
nde
nt
va
r
ia
bl
e
ve
c
to
r
.
A
ddi
ti
ona
ll
y,
you
m
us
t
s
pe
c
if
y
th
e
num
be
r
of
e
po
c
hs
,
w
hi
c
h
de
not
e
s
ho
w
m
a
n
y
ti
m
e
s
th
e
m
ode
l
w
il
l
it
e
r
a
te
ove
r
th
e
e
nt
ir
e
da
ta
s
e
t,
a
nd
th
e
ba
tc
h
s
iz
e
,
w
hi
c
h
in
di
c
a
te
s
th
e
num
be
r
of
tr
a
in
in
g
s
a
m
pl
e
s
pr
oc
e
s
s
e
d
be
f
or
e
th
e
m
ode
l'
s
in
te
r
na
l
pa
r
a
m
e
te
r
s
a
r
e
upda
te
d. W
it
h t
he
ANN
, t
h
e
pe
r
f
or
m
a
nc
e
s
um
m
a
r
iz
e
d i
n T
a
bl
e
4 w
a
s
obt
a
in
e
d.
T
a
bl
e
4. M
od
e
l
a
c
c
ur
a
c
y by
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
w
it
h
ANN
W
a
t
e
r
qua
l
i
t
y pa
r
a
m
e
t
e
r
P
r
e
di
c
t
or
va
r
i
a
bl
e
s
A
c
c
ur
a
c
y (
%
)
M
A
P
E
(
%
)
R
M
S
E
CL
-
a
'
r
_b1'
,'
r
_b3
'
, '
r
_b1_b2'
,
'
r
_b3_b4'
, '
r
_b2_b4'
, '
r
_ndvi
'
50
50
10.2 μ
g/
L
DO
'
r
_b4'
, '
r
_b3_b4
'
,'
r
_pc
1'
,'
r
_pc
4'
87
13
1.2 m
g O
2/
L
T
S
S
'
r
_b1'
,'
r
_b2
'
,'
r
_ndw
i
'
73
27
23.6 m
g/
L
S
a
l
i
ni
t
y
'
r
_b1'
, '
r
_b3
'
, '
r
_b2_b3'
,
'
r
_b3_b4'
, '
r
_b2_b4'
, '
r
_ndvi
'
,
'
r
_ndw
i
'
, '
r
_pc
1'
90
10
3.6 o/
oo
CL
-
a
w
a
s
th
e
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
w
it
h
th
e
lo
w
e
s
t
pe
r
f
or
m
a
nc
e
u
s
in
g
th
e
ANN
a
lg
or
it
hm
,
w
it
h
a
n
e
s
ti
m
a
te
d
a
c
c
ur
a
c
y
of
50%
.
T
hi
s
im
pl
ie
s
th
a
t
th
e
m
ode
l
a
c
c
ount
s
f
or
onl
y
50%
of
th
e
to
ta
l
va
r
ia
bi
li
ty
in
th
e
da
ta
.
A
lt
hough
hi
gh
pe
r
f
or
m
a
nc
e
w
a
s
not
a
c
hi
e
v
e
d
f
or
CL
-
a
,
RF
yi
e
ld
e
d
a
good
pe
r
f
or
m
a
nc
e
w
it
h
a
n
a
c
c
ur
a
c
y
of
79%
.
T
hi
s
f
a
ll
s
w
it
hi
n
th
e
pr
e
c
is
io
n
r
a
nge
r
e
por
te
d
in
s
tu
di
e
s
of
th
is
va
r
ia
bl
e
us
in
g
r
e
m
ot
e
s
e
ns
in
g
da
ta
,
w
hi
c
h
in
di
c
a
te
a
n
R
²
b
e
twe
e
n
0.5 a
nd
0.9
[
15]
.
A
ddi
ti
ona
ll
y,
A
gui
la
r
a
nd
D
ía
z
[
3]
r
e
por
te
d
th
a
t
AI
te
c
hni
que
s
c
a
n
a
c
hi
e
v
e
a
c
c
ur
a
c
ie
s
gr
e
a
t
e
r
th
a
n
60%
.
T
hi
s
r
e
s
ul
t
w
a
s
a
c
hi
e
ve
d
us
in
g
th
e
r
e
d
b
a
nd
a
nd
th
e
s
im
pl
e
quot
ie
nt
be
twe
e
n
th
e
bl
ue
a
nd
gr
e
e
n
b
a
nds
.
T
hi
s
a
li
gn
s
w
it
h
s
e
ve
r
a
l
s
tu
di
e
s
[
15]
,
[
19]
,
[
21]
,
a
m
ong
ot
he
r
s
,
th
a
t
a
s
s
oc
ia
te
th
e
vi
s
ib
le
s
pe
c
tr
um
w
it
h
th
e
w
a
t
e
r
qua
li
ty
pa
r
a
m
e
te
r
of
CL
-
a
.
P
hyt
opl
a
nkt
on
e
xhi
bi
ts
a
n
a
bs
or
pt
io
n
pe
a
k
in
th
e
bl
ue
b
a
nd
[
25]
,
w
it
h
w
a
ve
le
ngt
hs
b
e
twe
e
n
0.455
a
nd
0.515
µm
,
a
nd
a
r
e
f
le
c
ti
on
m
a
xi
m
um
in
th
e
gr
e
e
n
b
a
nd
,
w
it
h
w
a
ve
le
ngt
hs
be
twe
e
n
0
.5
a
nd
0.59
µm
,
f
ol
lo
w
e
d
by
a
de
c
li
ne
in
w
a
ve
le
ngt
hs
gr
e
a
te
r
t
ha
n 0.59 µm
.
DO
,
s
im
il
a
r
to
CL
-
a
,
de
m
ons
tr
a
te
d
it
s
hi
ghe
s
t
pe
r
f
or
m
a
nc
e
w
it
h
th
e
RF
a
lg
or
it
h
m
,
a
c
hi
e
vi
ng
a
n
e
s
ti
m
a
te
d
a
c
c
ur
a
c
y
of
95%
.
T
h
e
pr
e
di
c
to
r
va
r
ia
bl
e
s
f
or
th
is
m
ode
l
in
c
lu
de
th
e
ne
a
r
in
f
r
a
r
e
d
b
a
nd,
th
e
s
im
pl
e
quot
ie
nt
be
twe
e
n
th
e
gr
e
e
n
a
nd
ne
a
r
in
f
r
a
r
e
d
ba
nds
,
a
nd
m
a
in
c
om
pone
nt
4.
T
he
gr
e
e
n
ba
nd
is
pa
r
ti
c
ul
a
r
ly
not
e
w
or
th
y,
a
s
it
s
how
s
s
ig
ni
f
ic
a
nt
r
e
s
ul
ts
in
r
e
gr
e
s
s
io
n
m
ode
ls
,
s
uc
h
a
s
th
e
one
by
A
lz
a
te
e
t
al
.
[
26]
w
it
h
a
n
R
²
of
0.77, a
nd
th
e
s
im
pl
e
li
ne
a
r
r
e
gr
e
s
s
io
n
m
ode
l
f
r
om
V
a
ne
g
a
s
[
27]
w
it
h a
n
R
²
of
0.8.
T
he
r
e
f
or
e
,
th
i
s
r
e
s
ul
t
c
oul
d
be
va
lu
a
bl
e
f
or
pr
e
di
c
ti
ng
DO
le
ve
l
s
in
th
e
V
ir
ge
n
S
w
a
m
p
.
V
a
ne
ga
s
[
27]
,
in
hi
s
r
e
s
ul
ts
s
p
e
c
if
ie
d
th
a
t
he
di
d not f
in
d a
r
e
la
ti
ons
hi
p be
twe
e
n t
he
a
na
ly
z
e
d pa
r
a
m
e
te
r
s
a
nd t
he
ne
a
r
-
in
f
r
a
r
e
d, w
hi
c
h di
f
f
e
r
s
f
r
om
w
ha
t
is
pr
e
s
e
nt
e
d
he
r
e
s
in
c
e
th
i
s
ba
nd
c
ont
r
ib
ut
e
d
in
th
e
pr
e
di
c
ti
on,
not
onl
y
in
th
e
DO
w
he
r
e
a
S
pe
a
r
m
a
n
c
or
r
e
la
ti
on
c
oe
f
f
ic
ie
nt
of
0.3
w
a
s
obt
a
in
e
d
w
it
h
a
p
-
va
lu
e
o
f
0.018
lo
w
e
r
th
a
n
th
e
s
ta
ti
s
ti
c
a
l
s
ig
ni
f
ic
a
nc
e
α=
0.05,
w
hi
c
h
a
ll
ow
e
d
r
e
je
c
ti
ng
th
e
nul
l
hypothe
s
is
th
a
t
c
ons
id
e
r
s
th
a
t
th
e
ne
a
r
-
in
f
r
a
r
e
d
ba
nd
doe
s
no
t
c
or
r
e
la
te
w
it
h t
he
DO
pa
r
a
m
e
te
r
, but
a
ls
o c
ont
r
ib
ut
e
d i
n t
he
pr
e
di
c
ti
on of
t
he
ot
he
r
pa
r
a
m
e
te
r
s
s
tu
di
e
d.
T
S
S
de
m
ons
tr
a
te
d s
tr
ong pe
r
f
or
m
a
nc
e
a
c
r
os
s
a
ll
t
hr
e
e
pr
opos
e
d m
ode
ls
. T
he
m
os
t
s
ig
ni
f
ic
a
nt
r
e
s
ul
ts
w
e
r
e
a
c
hi
e
ve
d
w
it
h
th
e
RF
m
ode
l,
w
hi
c
h
ha
d
a
n
a
c
c
ur
a
c
y
of
89%
.
T
he
ke
y
pr
e
di
c
to
r
va
r
ia
bl
e
s
in
c
lu
de
d
th
e
bl
ue
a
nd
gr
e
e
n
ba
nds
,
a
s
w
e
ll
a
s
th
e
N
D
W
I
,
w
hi
c
h
is
de
r
iv
e
d
f
r
om
th
e
nor
m
a
li
z
e
d
di
f
f
e
r
e
nc
e
be
twe
e
n
th
e
gr
e
e
n
a
nd
N
I
R
ba
nds
.
T
he
s
e
ba
nd
s
w
e
r
e
a
l
s
o
c
r
uc
ia
l
f
or
pr
e
di
c
ti
ng
T
S
S
in
th
e
s
tu
dy
by
[
28]
,
w
he
r
e
th
e
be
s
t
pr
e
di
c
ti
ons
w
e
r
e
obt
a
in
e
d
u
s
in
g
th
e
quot
ie
nt
of
th
e
s
e
va
r
ia
bl
e
s
.
T
hi
s
m
ode
l
a
ls
o
s
how
e
d
th
e
s
m
a
ll
e
s
t
m
e
a
n
s
qua
r
e
d
e
r
r
or
(
M
S
E
)
a
nd
M
A
P
E
,
c
ons
is
te
nt
w
it
h
th
e
m
e
tr
ic
s
us
e
d
in
th
is
p
a
pe
r
.
T
h
e
s
e
ba
nds
h
a
ve
be
e
n
e
m
pl
oye
d
in
va
r
io
us
s
tu
di
e
s
to
de
te
r
m
in
e
T
S
S
,
of
te
n
in
c
onj
unc
ti
on
w
it
h
s
im
pl
e
quot
ie
nt
s
.
F
or
in
s
ta
nc
e
,
G
hol
iz
a
de
h
e
t
al
.
[
29]
ut
il
iz
e
d
th
e
s
e
ba
nds
w
it
h
m
ode
l
s
in
c
o
r
por
a
ti
ng
DL
te
c
hni
que
s
,
w
hi
le
[
30]
a
ppl
ie
d
s
im
pl
e
r
e
gr
e
s
s
io
ns
u
s
in
g
th
e
vi
s
ib
le
s
pe
c
tr
um
ba
nds
f
r
om
th
e
T
M
s
e
ns
or
on
th
e
L
a
nds
a
t
5
pl
a
tf
or
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
v
al
uat
io
n of
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
al
gor
it
hm
s
t
o e
s
ti
m
at
e
w
at
e
r
quali
ty
…
(
J
ul
io
C
e
s
a
r
A
nay
a
-
V
al
e
nz
ue
la
)
565
H
ow
e
ve
r
,
th
e
la
tt
e
r
r
e
f
e
r
e
nc
e
r
e
por
te
d
a
lo
w
pe
r
f
or
m
a
nc
e
f
or
th
e
ir
m
ode
l.
RF
ha
ve
c
ont
r
ib
ut
e
d
w
it
h
s
a
ti
s
f
a
c
to
r
y
r
e
s
ul
ts
in
ot
he
r
s
tu
di
e
s
f
or
oc
c
upa
ti
ona
l
s
a
f
e
ty
a
nd
he
a
lt
h
(
O
S
H
)
m
ode
ll
in
g
s
uc
h
a
s
th
e
one
pr
e
s
e
nt
e
d by
[
31]
, a
lt
hough the
c
or
r
e
la
ti
on i
s
m
a
de
w
it
h ot
he
r
phys
ic
oc
he
m
ic
a
l
pa
r
a
m
e
te
r
s
of
w
a
te
r
qua
li
ty
.
T
he
be
s
t
m
ode
l
f
or
r
e
pr
e
s
e
nt
in
g
s
a
li
ni
ty
w
a
s
th
e
RF
a
lg
or
it
hm
,
a
c
hi
e
vi
ng
a
n
a
c
c
ur
a
c
y
of
96%
.
T
hi
s
m
ode
l
out
pe
r
f
or
m
e
d
th
e
S
V
M
m
ode
ls
,
w
hi
c
h
ha
d
a
n
a
c
c
ur
a
c
y
of
94%
,
a
nd
th
e
ANN
,
w
hi
c
h
a
c
hi
e
ve
d
a
n
a
c
c
ur
a
c
y of
90%
. T
he
l
a
tt
e
r
s
how
e
d f
a
vor
a
bl
e
r
e
s
ul
t
s
c
om
pa
r
e
d
t
o
m
ul
ti
pl
e
l
in
e
a
r
r
e
gr
e
s
s
io
n
m
ode
ls
, a
s
not
e
d
by
Z
hou
e
t
al
.
[
32]
,
a
nd
de
m
ons
tr
a
te
d
a
c
c
ur
a
c
ie
s
e
qua
l
to
o
r
e
xc
e
e
di
ng
90%
,
a
s
r
e
por
te
d
in
th
e
s
a
li
ni
ty
va
r
ia
ti
on
s
tu
dy
by
H
ua
ng
a
nd
F
oo
[
33]
.
S
a
li
ni
ty
w
a
s
th
e
pa
r
a
m
e
te
r
w
it
h
th
e
be
s
t
pe
r
f
or
m
a
nc
e
th
r
oughout
th
e
a
na
ly
s
is
.
T
hi
s
w
a
s
e
vi
de
nc
e
d
by
a
S
pe
a
r
m
a
n
c
or
r
e
la
ti
on
c
oe
f
f
ic
ie
nt
of
0.6
a
nd
a
p
-
va
lu
e
of
le
s
s
th
a
n
0.05,
w
hi
c
h
w
e
r
e
obt
a
in
e
d
f
r
om
th
e
di
r
e
c
t
a
n
a
ly
s
is
of
th
e
s
im
pl
e
quot
ie
nt
be
twe
e
n
th
e
bl
ue
b
a
nd
(
0.455
to
0.515 µm
)
a
nd t
he
gr
e
e
n b
a
nd (
0.5 t
o 0.59 µm
)
.
N
um
e
r
ic
a
l
m
ode
li
ng
w
it
h
in
f
or
m
a
ti
on
a
c
qui
r
e
d
th
r
ough
r
e
m
ot
e
s
e
ns
in
g
ha
s
de
m
ons
tr
a
te
d
it
s
us
e
f
ul
ne
s
s
in
pr
e
di
c
ti
ng
w
a
te
r
qua
li
ty
pa
r
a
m
e
te
r
s
[
21]
s
uc
h
a
s
th
os
e
s
e
le
c
te
d
in
th
is
s
tu
dy.
T
he
pe
r
f
or
m
a
nc
e
of
th
e
m
ode
ls
is
in
f
lu
e
nc
e
d
by
f
a
c
to
r
s
s
uc
h
a
s
th
e
ti
m
e
di
f
f
e
r
e
nc
e
be
twe
e
n
w
he
n
th
e
la
bor
a
to
r
y
s
a
m
pl
e
is
c
ol
le
c
te
d
a
nd
w
h
e
n
th
e
r
e
m
ot
e
s
e
ns
or
im
a
ge
i
s
c
a
pt
ur
e
d.
T
he
r
e
f
or
e
,
it
is
c
r
uc
ia
l
to
e
ns
ur
e
th
a
t
th
e
da
ta
a
r
e
c
ol
le
c
te
d
a
t
th
e
s
a
m
e
ti
m
e
.
A
s
not
e
d
by
B
a
z
á
n
e
t
al
.
[
21]
,
if
th
e
r
e
a
r
e
te
m
por
a
l
di
s
c
r
e
p
a
nc
ie
s
,
it
is
e
s
s
e
nt
ia
l
to
ve
r
if
y
th
a
t
th
e
w
a
te
r
body
ha
s
not
e
xpe
r
ie
nc
e
d
di
s
tu
r
ba
nc
e
s
,
s
u
c
h
a
s
pr
e
c
ip
it
a
ti
on
or
c
ont
a
m
in
a
nt
di
s
c
ha
r
ge
s
,
th
a
t
c
oul
d
a
lt
e
r
th
e
s
pe
c
tr
a
l
r
e
s
pons
e
of
th
e
s
a
te
ll
it
e
im
a
ge
or
in
tr
oduc
e
a
nom
a
lo
us
va
lu
e
s
or
out
li
e
r
s
in
th
e
la
bor
a
to
r
y
s
a
m
pl
e
.
T
he
s
e
out
li
e
r
s
s
houl
d
be
e
xc
lu
de
d
f
r
om
t
he
s
a
m
pl
e
to
a
voi
d
ne
ga
ti
ve
ly
im
pa
c
ti
ng
th
e
a
lg
or
it
hm
'
s
pe
r
f
or
m
a
nc
e
.
I
t
is
im
po
r
ta
nt
to
a
c
knowle
dge
th
e
li
m
it
a
ti
ons
of
th
is
s
tu
dy,
pr
im
a
r
il
y
th
e
s
m
a
ll
s
a
m
pl
e
s
iz
e
.
T
he
da
ta
s
e
t
c
on
s
is
te
d
of
onl
y
37
s
a
m
pl
in
g
r
e
c
or
ds
,
w
hi
c
h
c
ons
tr
a
in
e
d
th
e
m
ode
l
tr
a
in
in
g
a
nd
va
li
da
ti
on
pr
oc
e
s
s
,
e
s
p
e
c
ia
ll
y
f
or
c
om
pl
e
x a
lg
or
it
hm
s
li
ke
A
N
N
.
C
ons
e
qu
e
nt
ly
,
th
e
va
li
d
a
ti
on
w
a
s
pe
r
f
or
m
e
d
us
in
g
a
s
im
pl
e
70/
30
tr
a
in
-
te
s
t
s
pl
it
.
W
hi
le
th
is
pr
ovi
de
d
in
it
i
a
l
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
th
e
r
obus
tn
e
s
s
of
th
e
m
ode
ls
c
oul
d
be
f
ur
th
e
r
im
pr
ove
d
by
e
m
pl
oyi
ng
m
or
e
r
ig
or
ou
s
te
c
hni
que
s
,
s
uc
h
a
s
k
-
f
ol
d
c
r
os
s
-
va
li
da
ti
on,
w
hi
c
h
is
be
tt
e
r
s
ui
te
d
f
or
li
m
it
e
d
da
ta
s
e
t
s
.
F
ut
ur
e
w
or
k
s
houl
d
a
im
to
in
c
or
por
a
te
a
la
r
ge
r
da
ta
s
e
t
to
va
li
da
te
a
nd e
nha
nc
e
t
he
g
e
ne
r
a
li
z
a
bi
li
ty
of
t
he
s
e
f
in
di
ngs
.
F
or
f
ut
ur
e
r
e
s
e
a
r
c
h,
e
xpl
or
in
g
m
or
e
a
dva
nc
e
d
DL
a
r
c
hi
te
c
tu
r
e
s
c
oul
d
yi
e
ld
s
ig
ni
f
ic
a
nt
im
pr
ove
m
e
nt
s
.
F
or
in
s
ta
nc
e
,
t
r
a
ns
f
or
m
e
r
-
ba
s
e
d
de
e
p
ne
twor
ks
,
w
hi
c
h
ha
ve
s
how
n
s
uc
c
e
s
s
in
noi
s
e
r
e
duc
ti
on
f
or
ot
he
r
dom
a
in
s
li
ke
m
e
di
c
a
l
im
a
gi
ng,
c
oul
d
be
a
da
pt
e
d
f
or
pr
oc
e
s
s
in
g
s
a
te
ll
it
e
im
a
ge
r
y.
S
uc
h
m
ode
ls
m
ig
ht
pr
ove
e
f
f
e
c
ti
ve
in
m
it
ig
a
ti
ng
a
tm
os
phe
r
ic
in
te
r
f
e
r
e
nc
e
a
n
d
ot
he
r
noi
s
e
in
he
r
e
nt
in
r
e
m
ot
e
s
e
ns
in
g
da
ta
,
th
e
r
e
by i
m
pr
ovi
ng t
he
qua
li
ty
of
s
pe
c
tr
a
l
s
ig
na
tu
r
e
s
us
e
d f
or
w
a
te
r
qua
li
ty
e
s
ti
m
a
ti
on.
4.
C
O
N
C
L
U
S
I
O
N
T
he
u
s
e
of
s
a
te
ll
it
e
im
a
ge
r
y
c
om
bi
ne
d
w
it
h
AI
m
od
e
ls
c
ons
ti
t
ut
e
s
a
n
e
f
f
e
c
ti
ve
c
om
pl
e
m
e
nt
a
r
y
to
ol
f
or
m
oni
to
r
in
g
w
a
te
r
qua
li
ty
in
le
nt
ic
bodi
e
s
s
uc
h
a
s
th
e
V
ir
ge
n
S
w
a
m
p,
a
s
it
r
e
ve
a
l
s
s
ig
ni
f
ic
a
nt
r
e
la
ti
ons
hi
ps
be
twe
e
n
s
pe
c
tr
a
l
r
e
f
le
c
ta
nc
e
a
nd
pa
r
a
m
e
te
r
s
s
uc
h
a
s
C
L
-
a
,
D
O
,
s
a
li
ni
ty
,
a
nd
T
S
S
,
a
lt
hough
it
doe
s
not
r
e
pl
a
c
e
la
bor
a
to
r
y
a
na
ly
s
e
s
.
ML
a
nd
DL
m
ode
l
s
s
ho
w
e
d
a
dv
a
n
ta
ge
s
ove
r
tr
a
di
ti
ona
l
s
ta
ti
s
ti
c
a
l
a
ppr
oa
c
he
s
by
not
r
e
ly
in
g
on
pa
r
a
m
e
tr
ic
a
s
s
um
pt
io
ns
a
nd
by
a
c
hi
e
vi
ng
hi
gh
l
e
ve
ls
of
a
c
c
ur
a
c
y,
w
it
h
s
a
li
ni
ty
s
ta
ndi
ng
out
a
s
th
e
be
s
t
-
pe
r
f
or
m
in
g
pa
r
a
m
e
te
r
(
up
to
96%
w
it
h
RF
)
,
w
hi
le
C
L
-
a
pr
e
s
e
nt
e
d
th
e
lo
w
e
s
t
pe
r
f
or
m
a
nc
e
,
pos
s
ib
ly
due
to
th
e
in
f
lu
e
nc
e
of
th
e
s
w
a
m
p
be
d.
O
ve
r
a
ll
,
th
e
r
e
s
ul
ts
c
onf
ir
m
th
a
t
th
e
s
e
m
ode
ls
e
na
bl
e
e
f
f
ic
ie
nt
,
lo
w
-
c
os
t
m
oni
to
r
in
g
w
it
h
pot
e
nt
ia
l
a
s
a
n
e
a
r
ly
w
a
r
ni
ng
s
ys
te
m
f
or
e
nvi
r
onm
e
nt
a
l
m
a
na
ge
m
e
nt
,
s
uppor
te
d
by
ope
n
pl
a
tf
or
m
s
s
uc
h
a
s
G
oogl
e
E
a
r
th
E
ngi
ne
a
nd
J
upyt
e
r
N
ot
e
book,
a
nd
pr
ovi
de
va
lu
a
bl
e
in
f
or
m
a
ti
on
f
or
de
c
is
io
n
-
m
a
ki
ng by e
nvi
r
onm
e
nt
a
l
a
ut
hor
it
ie
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
J
ul
io
C
e
s
a
r
A
na
y
a
-
V
a
le
nz
ue
la
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
G
lo
r
ia
Y
a
ne
th
F
lo
r
e
z
-
Y
e
pe
s
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Y
e
is
on A
l
be
r
to
G
a
r
c
é
s
-
G
óm
e
z
✓
✓
✓
✓
✓
✓
✓
✓
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
15
, N
o.
1
,
F
e
br
ua
r
y
20
26
:
559
-
567
566
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
t
a
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
a
va
il
a
bl
e
f
r
om
th
e
c
or
r
e
s
ponding
a
ut
hor
,
[
Y
A
G
G
]
, upon
r
e
a
s
ona
bl
e
r
e
que
s
t.
R
E
F
E
R
E
N
C
E
S
[
1]
E
.
C
huvi
e
c
o,
F
undam
e
nt
al
s
of
s
at
e
l
l
i
t
e
r
e
m
ot
e
s
e
ns
i
ng
an
e
nv
i
r
onm
e
nt
al
appr
oac
h
.
N
e
w
Y
or
k,
U
ni
t
e
d
S
t
a
t
e
s
:
C
R
C
P
r
e
s
s
,
T
a
yl
or
&
F
r
a
nc
i
s
G
r
oup, 2020.
[
2]
J
.
A
.
-
A
l
va
r
e
z
,
P
.
P
.
-
C
ut
i
l
l
a
s
,
L
.
C
.
A
.
-
C
e
j
udo,
a
nd
O
.
R
.
-
V
a
l
l
e
,
“
M
ul
t
i
s
pe
c
t
r
a
l
a
na
l
ys
i
s
f
or
e
s
t
i
m
a
t
i
ng
t
ur
bi
di
t
y
a
s
a
n
i
ndi
c
a
t
or
of
w
a
t
e
r
qua
l
i
t
y
i
n
r
e
s
e
r
voi
r
s
i
n
t
he
s
t
a
t
e
of
C
hi
hua
hu
a
,
M
e
xi
c
o
(
i
n
S
pa
ni
s
h:
A
nál
i
s
i
s
m
ul
t
i
e
s
pe
c
t
r
al
par
a
l
a
e
s
t
i
m
ac
i
ón
de
l
a
t
ur
bi
de
z
c
om
o
i
ndi
c
ado
r
de
l
a
c
al
i
dad
de
l
agua
e
n
e
m
bal
s
e
s
de
l
e
s
t
ado
de
C
hi
huahua,
M
é
x
i
c
o
)
,”
R
e
v
i
s
t
a
G
e
ogr
áf
i
c
a
de
A
m
é
r
i
c
a
C
e
nt
r
al
, vol
. 1, no. 62, pp. 33
–
61, S
e
p. 2018, doi
:
10.15359/
r
ga
c
.62
-
1.2.
[
3]
A
. C
. A
.
A
gui
l
a
r
a
nd F
. F
. O
.
-
D
í
a
z
, “
M
a
c
hi
ne
l
e
a
r
ni
ng f
or
pr
e
di
c
t
i
ng dr
i
nki
ng w
a
t
e
r
qua
l
i
t
y,”
I
nge
ni
ar
e
, no.
28, pp. 47
–
62,
2020,
doi
:
10.18041/
1909
-
2458/
i
nge
ni
a
r
e
.28.6215.
[
4]
W
.
M
a
l
dona
do,
I
.
B
a
l
di
r
i
s
,
a
nd
J
.
D
í
a
z
,
“
A
s
s
e
s
s
m
e
nt
of
w
a
t
e
r
qua
l
i
t
y
of
C
i
é
n
a
ga
de
l
a
V
i
r
ge
n
(
C
a
r
t
a
ge
na
,
C
ol
om
bi
a
)
dur
i
ng
t
he
pe
r
i
od 2006
-
2010,”
R
e
v
i
s
t
a C
i
e
nt
í
f
i
c
a G
ui
l
l
e
r
m
o de
O
c
k
ham
, vol
. 9, no. 2, pp.
79
–
87, D
e
c
. 2011.
[
5]
C
a
r
di
que
,
P
l
an
f
or
t
he
m
anage
m
e
nt
and
c
ons
e
r
v
at
i
on
of
t
he
C
i
e
naga
de
l
a
V
i
r
ge
n
w
at
e
r
s
he
d
(
i
n
Spani
s
h:
P
l
an
de
O
r
de
na
m
i
e
nt
o
y
M
ane
j
o
de
l
a
C
ue
nc
a
H
i
d
r
ogr
af
i
c
a
de
l
a
C
i
e
naga
de
l
a
V
i
r
ge
n
)
.
B
ogot
á
,
C
ol
om
bi
a
:
C
a
r
di
que
-
C
ons
e
r
va
c
i
on
I
nt
e
r
na
t
i
ona
l
C
ol
om
bi
a
, 2019. [
O
nl
i
ne
]
. A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
e
s
.
s
c
r
i
bd.c
om
/
doc
/
7096498/
01
-
P
l
a
n
-
to
-
y
-
M
a
ne
j
o
-
C
ue
nc
a
-
C
i
e
na
ga
-
de
-
La
-
V
i
r
ge
n
[
6]
R
.
H
ua
ng,
C
.
M
a
,
J
.
M
a
,
X
.
H
ua
ngf
u,
a
nd
Q
.
H
e
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
i
n
na
t
ur
a
l
a
nd
e
ngi
ne
e
r
e
d
w
a
t
e
r
s
ys
t
e
m
s
,”
W
at
e
r
R
e
s
e
ar
c
h
,
vol
. 205, O
c
t
. 2021, doi
:
10.1016/
j
.w
a
t
r
e
s
.2021.117666.
[
7]
A
.
G
i
t
e
l
s
on,
G
.
G
a
r
buz
ov,
F
.
S
z
i
l
a
gyi
,
K
.
H
.
M
i
t
t
e
nz
w
e
y,
A
.
K
a
r
ni
e
l
i
,
a
nd
A
.
K
a
i
s
e
r
,
“
Q
ua
nt
i
t
a
t
i
ve
r
e
m
ot
e
s
e
ns
i
ng
m
e
t
hod
s
f
or
r
e
a
l
-
t
i
m
e
m
oni
t
or
i
ng
of
i
n
l
a
nd
w
a
t
e
r
s
qua
l
i
t
y,”
I
nt
e
r
nat
i
onal
J
our
nal
of
R
e
m
ot
e
Se
ns
i
ng
,
vol
.
14,
no.
7,
pp.
1269
–
1295,
1993,
doi
:
10.1080/
01431169308953956.
[
8]
R
. B
a
i
r
d a
nd L
. B
r
i
dge
w
a
t
e
r
,
S
t
and
ar
d m
e
t
hods
f
or
t
he
e
x
am
i
nat
i
on of
w
at
e
r
an
d w
as
t
e
w
at
e
r
s
t
andar
d m
e
t
hods
f
or
t
he
e
x
am
i
na
t
i
on
of
w
at
e
r
an
d w
as
t
e
w
at
e
r
,
23r
d e
d.,
no.
1. W
a
s
hi
n
gt
on
, D
. C
,
U
ni
t
e
d S
t
a
t
e
s
:
A
m
e
r
i
c
a
n
P
ubl
i
c
H
e
a
l
t
h A
s
s
oc
i
a
t
i
on,
2017.
[
9]
O
bs
e
r
va
t
or
i
o
A
m
bi
e
nt
a
l
D
e
C
a
r
t
a
ge
na
de
I
ndi
a
s
,
“
C
i
é
na
ga
de
L
a
V
i
r
ge
n,”
obs
e
r
v
at
or
i
o.e
pa
c
ar
t
age
na.gov
.c
o
.
A
c
c
e
s
s
e
d:
J
a
n.
20,
2025.
[
O
nl
i
ne
]
.
A
v
a
i
l
a
bl
e
:
ht
t
ps
:
/
/
obs
e
r
va
t
or
i
o.e
pa
c
a
r
t
a
ge
na
.gov.c
o/
ge
s
t
i
on
-
a
m
bi
e
nt
a
l
/
e
c
os
i
s
t
e
m
a
s
/
pr
oye
c
t
o
-
c
i
e
na
g
a
-
de
-
la
-
vi
r
ge
n/
c
i
e
na
ga
-
de
-
la
-
vi
r
ge
n/
[
10]
Y
.
O
ya
m
a
,
B
.
M
a
t
s
us
hi
t
a
,
T
.
F
ukus
hi
m
a
,
K
.
M
a
t
s
u
s
hi
ge
,
a
nd
A
.
I
m
a
i
,
“
A
p
pl
i
c
a
t
i
on
of
s
pe
c
t
r
a
l
de
c
om
po
s
i
t
i
on
a
l
gor
i
t
hm
f
or
m
a
ppi
ng
w
a
t
e
r
qua
l
i
t
y
i
n
a
t
ur
bi
d
l
a
ke
(
L
a
ke
K
a
s
um
i
ga
ur
a
,
J
a
pa
n)
f
r
o
m
L
a
nds
a
t
T
M
da
t
a
,”
I
SP
R
S
J
our
nal
of
P
hot
ogr
am
m
e
t
r
y
and R
e
m
ot
e
Se
ns
i
ng
, vol
. 64, no. 1, pp. 73
–
85, J
a
n. 2009, doi
:
10.1016/
j
.i
s
pr
s
j
pr
s
.2008.04.005.
[
11]
R
. M
. M
c
C
oy,
F
i
e
l
d m
e
t
hods
i
n r
e
m
ot
e
s
e
n
s
i
ng
. N
e
w
Y
or
k, U
ni
t
e
d S
t
a
t
e
s
:
T
he
G
ui
l
f
or
d P
r
e
s
s
, 2005.
[
12]
P
l
a
ne
t
,
“
P
l
a
ne
t
pr
oduc
t
s
:
r
e
a
l
-
t
i
m
e
s
a
t
e
l
l
i
t
e
m
oni
t
or
i
ng
w
i
t
h
pl
a
ne
t
,”
pl
ane
t
.c
om
.
A
c
c
e
s
s
e
d:
J
a
n.
20,
2025.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.pl
a
ne
t
.c
om
/
pr
oduc
t
s
/
s
a
t
e
l
l
i
t
e
-
m
oni
t
or
i
ng/
[
13]
O
.
O
.
D
i
a
z
,
B
as
i
c
i
nt
r
oduc
t
i
on
t
o
G
oogl
e
E
ar
t
h
E
ngi
ne
(
G
E
E
)
(
i
n
Spani
s
h:
I
n
t
r
oduc
c
i
ón
bás
i
c
a
a
G
oogl
e
E
ar
t
h
e
ngi
ne
(
G
E
E
)
)
.
B
onn, G
e
r
m
a
ny:
D
e
ut
s
c
he
G
e
s
e
l
l
s
c
ha
f
t
f
ür
I
nt
e
r
na
t
i
ona
l
e
Z
us
a
m
m
e
na
r
be
i
t
(
G
I
Z
)
, 2018.
[
14]
J
.
X
i
ong
e
t
al
.
,
“
N
om
i
na
l
30
-
m
c
r
opl
a
nd
e
xt
e
nt
m
a
p
of
c
ont
i
ne
nt
a
l
A
f
r
i
c
a
by
i
nt
e
gr
a
t
i
ng
pi
xe
l
-
ba
s
e
d
a
nd
obj
e
c
t
-
ba
s
e
d
a
l
gor
i
t
hm
s
us
i
ng S
e
nt
i
ne
l
-
2 a
nd L
a
nds
a
t
-
8 da
t
a
on G
oogl
e
E
a
r
t
h E
ngi
ne
,”
R
e
m
ot
e
Se
ns
i
ng
, vol
. 9, no. 10, O
c
t
. 2017, doi
:
10.3390/
r
s
9101065.
[
15]
I
.
B
r
i
c
e
ño,
W
.
P
é
r
e
z
,
D
.
S
.
M
i
gue
l
,
a
nd
S
.
R
a
m
os
,
“
D
e
t
e
r
m
i
na
t
i
on
of
w
a
t
e
r
qua
l
i
t
y
V
i
c
huqué
n
L
a
ke
,
us
i
ng
s
a
t
e
l
l
i
t
e
i
m
a
ge
s
L
a
nds
a
t
8,
s
e
ns
or
O
L
I
, ye
a
r
2016, C
hi
l
e
,”
R
e
v
i
s
t
a de
T
e
l
e
d
e
t
e
c
c
i
ón
, no. 52, pp.
67
–
78, 2018, doi
:
10.4995/
r
a
e
t
.2018.10126.
[
16]
C
.
P
ohl
e
t
al
.
,
P
r
i
nc
i
pl
e
s
of
r
e
m
ot
e
s
e
ns
i
ng
an
i
nt
r
oduc
t
or
y
t
e
x
t
book
.
E
ns
c
he
de
,
N
e
t
he
r
l
a
nd:
T
he
I
nt
e
r
na
t
i
ona
l
I
ns
t
i
t
ut
e
f
or
A
e
r
os
pa
c
e
S
ur
ve
y a
nd E
a
r
t
h S
c
i
e
nc
e
s
(
I
T
C
)
, 2001.
[
17]
K
.
G
.
R
uddi
c
k,
V
.
D
e
C
a
uw
e
r
,
Y
.
J
.
P
a
r
k,
a
nd
G
.
M
oor
e
,
“
S
e
a
bor
ne
m
e
a
s
ur
e
m
e
nt
s
of
ne
a
r
i
nf
r
a
r
e
d
w
a
t
e
r
-
l
e
a
vi
ng
r
e
f
l
e
c
t
a
nc
e
:
t
he
s
i
m
i
l
a
r
i
t
y
s
pe
c
t
r
um
f
o
r
t
u
r
bi
d
w
a
t
e
r
s
,”
L
i
m
nol
ogy
and
O
c
e
anogr
aphy
,
vol
.
51,
no.
2,
pp.
1167
–
1179,
2006,
doi
:
10.4319/
l
o.2006.51.2.1167.
[
18]
J
.
T
.
-
P
é
r
e
z
a
n
d
A
.
M
c
C
u
l
l
um
,
“
R
e
m
o
t
e
s
e
ns
i
ng
of
c
oa
s
t
a
l
e
c
os
ys
t
e
m
s
,”
N
A
S
A
A
pp
l
i
e
d
R
e
m
o
t
e
S
e
ns
i
n
g
T
r
a
i
ni
ng
P
r
og
r
a
m
(
A
R
S
E
T
)
.
A
c
c
e
s
s
e
d
:
J
a
n. 20
, 2020.
[
O
nl
i
ne
.]
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
a
ppl
i
e
ds
c
i
e
nc
e
s
.
na
s
a
.gov/
r
e
m
ot
e
-
s
e
ns
i
ng
-
c
oa
s
t
a
l
-
e
c
os
y
s
t
e
m
s
[
19]
M
.
B
ona
ns
e
a
,
C
.
L
e
de
s
m
a
,
C
.
R
odr
i
gue
z
,
a
nd
A
.
R
.
S
.
D
e
l
ga
do,
“
C
hl
or
ophyl
l
-
a
c
onc
e
nt
r
a
t
i
on
a
nd
phot
i
c
z
one
bounda
r
y
i
n
t
he
R
í
o
T
e
r
c
e
r
o
r
e
s
e
r
voi
r
(
A
r
ge
nt
i
na
)
us
i
ng
C
B
E
R
S
-
2B
s
a
t
e
l
l
i
t
e
i
m
a
ge
s
(
i
n
S
pa
ni
s
h:
C
onc
e
nt
r
ac
i
ón
de
c
l
o
r
of
i
l
a
-
a
y
l
í
m
i
t
e
de
z
ona
f
ót
i
c
a
e
n
e
l
e
m
bal
s
e
R
í
o
T
e
r
c
e
r
o
(
A
r
ge
nt
i
na)
ut
i
l
i
z
ando
i
m
áge
n
e
s
d
e
l
s
at
é
l
i
t
e
C
B
E
R
S
-
2B
)
,”
A
m
bi
e
nt
e
e
A
gua
-
A
n
I
nt
e
r
di
s
c
i
pl
i
nar
y
J
ou
r
nal
of
A
ppl
i
e
d Sc
i
e
nc
e
, vol
. 7, no. 3, pp. 61
–
71, D
e
c
. 201
2, doi
:
10.4136/
a
m
bi
-
a
gua
.847.
[
20]
J
.
W
.
B
.
L
ope
s
,
F
.
B
.
L
ope
s
,
E
.
M
.
de
A
ndr
a
de
,
L
.
C
.
G
.
C
ha
v
e
s
,
a
nd
M
.
G
.
R
.
C
a
r
ne
i
r
o,
“
S
pe
c
t
r
a
l
r
e
s
pons
e
of
w
a
t
e
r
unde
r
di
f
f
e
r
e
nt
c
onc
e
nt
r
a
t
i
ons
of
s
us
pe
nde
d
s
e
di
m
e
nt
:
m
e
a
s
ur
e
m
e
nt
a
nd
s
i
m
pl
i
f
i
e
d
m
ode
l
i
n
g,”
J
our
nal
of
A
gr
i
c
ul
t
ur
al
Sc
i
e
nc
e
,
vol
. 11, no. 3, F
e
b. 2019, doi
:
10.5539/
j
a
s
.v11n3p327.
[
21]
R
.
B
a
z
á
n
e
t
al
.
,
“
R
e
m
ot
e
s
e
n
s
i
ng
a
nd
num
e
r
i
c
a
l
m
ode
l
i
ng
f
or
w
a
t
e
r
qu
a
l
i
t
y
a
na
l
ys
i
s
of
t
h
e
L
o
s
M
ol
i
nos
r
e
s
e
r
voi
r
,”
I
nge
ni
e
r
í
a
hi
dr
ául
i
c
a e
n M
é
x
i
c
o
, vol
. 20, no. 2, pp. 121
–
136, 2005.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
E
v
al
uat
io
n of
ar
ti
fi
c
ia
l
in
te
ll
ig
e
nc
e
al
gor
it
hm
s
t
o e
s
ti
m
at
e
w
at
e
r
quali
ty
…
(
J
ul
io
C
e
s
a
r
A
nay
a
-
V
al
e
nz
ue
la
)
567
[
22]
J
.
W
.
M
.
C
.
S
a
nt
o
s
a
nd
V
.
D
ubr
e
ui
l
,
“
E
s
t
i
m
a
t
i
on
of
t
he
t
e
m
por
a
l
a
nd
s
pa
t
i
a
l
di
s
t
r
i
but
i
on
of
s
us
pe
nde
d
m
a
t
e
r
i
a
l
i
n
t
h
e
w
a
t
e
r
s
of
t
he
M
a
ns
o
-
M
T
r
e
s
e
r
voi
r
ba
s
e
d
on
L
a
nds
a
t
i
m
a
ge
s
a
nd
f
i
e
l
d
da
t
a
(
i
n
P
or
t
ugue
s
e
:
E
s
t
i
m
at
i
v
a
da
di
s
t
r
i
bui
ç
ão
t
e
m
por
o
-
e
s
pac
i
al
de
m
at
e
r
i
al
e
m
s
us
p
e
ns
ão
nas
água
s
do
r
e
s
e
r
v
at
ó
r
i
o
de
m
ans
o
-
m
t
a
par
t
i
r
de
i
m
age
ns
l
ands
at
e
dados
de
c
am
po
)
,”
A
nai
s
X
I
V
Si
m
pos
i
o B
r
as
i
l
e
i
r
o de
S
e
ns
or
i
am
e
nt
o R
e
m
ot
o, N
at
al
, B
r
as
i
l
, pp. 5421
–
5428, 2
009.
[
23]
F
.
P
e
dr
e
gos
a
,
G
.
V
a
r
oqua
ux,
A
.
G
r
a
m
f
or
t
,
a
nd
V
.
M
i
c
he
l
,
“
H
i
s
t
or
y,”
s
c
i
k
i
t
-
l
e
ar
n.or
g
.
A
c
c
e
s
s
e
d:
J
a
n.
20,
2025.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
s
c
i
ki
t
-
l
e
a
r
n.or
g/
s
t
a
bl
e
/
a
bout
.ht
m
l
[
24]
C
.
L
e
de
s
m
a
,
M
.
B
ona
n
s
e
a
,
C
.
R
odr
í
gue
z
,
a
nd
Á
.
R
.
S
.
D
e
l
ga
do,
“
W
a
t
e
r
qua
l
i
t
y
c
ont
r
ol
i
n
t
hi
r
d
r
i
ve
r
r
e
s
e
r
voi
r
(
A
r
ge
nt
i
na
)
us
i
n
g
ge
ogr
a
phi
c
a
l
i
nf
or
m
a
t
i
on
s
ys
t
e
m
s
a
nd
l
i
ne
a
r
r
e
gr
e
s
s
i
on
m
ode
l
s
,
”
A
m
bi
e
nt
e
e
A
gua
-
A
n
I
nt
e
r
di
s
c
i
pl
i
na
r
y
J
ou
r
nal
of
A
ppl
i
e
d
Sc
i
e
nc
e
, vol
. 8, no. 2, pp. 67
–
76, 2013, doi
:
10.4136/
a
m
bi
-
a
gua
.1113.
[
25]
J
.
P
.
C
a
nni
z
z
a
r
o
a
nd
K
.
L
.
C
a
r
de
r
,
“
E
s
t
i
m
a
t
i
ng
c
hl
or
ophyl
l
a
c
onc
e
nt
r
a
t
i
ons
f
r
om
r
e
m
ot
e
-
s
e
ns
i
ng
r
e
f
l
e
c
t
a
nc
e
i
n
opt
i
c
a
l
l
y
s
h
a
l
l
ow
w
a
t
e
r
s
,”
R
e
m
ot
e
Se
ns
i
ng of
E
nv
i
r
onm
e
nt
, vol
. 101, no. 1, pp. 13
–
24, M
a
r
. 2006
, doi
:
10.1016/
j
.r
s
e
.2005.12.002.
[
26]
D
. F
. C
.
-
A
l
z
a
t
e
,
Y
. A
. G
.
-
G
o
m
e
z
, a
nd V
. H
.
-
C
e
s
pe
de
s
, “
L
a
nds
a
t
-
7
E
T
M
+ ba
s
e
d r
e
m
o
t
e
s
e
ns
i
ng a
s
a
t
o
ol
f
or
a
s
s
e
s
s
i
ng l
a
ke
s
w
a
t
e
r
qua
l
i
t
y c
ha
r
a
c
t
e
r
i
s
t
i
c
s
,”
J
o
ur
nal
o
f
Sou
t
hw
e
s
t
J
i
aot
o
ng U
n
i
v
e
r
s
i
t
y
, vo
l
. 5
6, no
. 1,
2021,
doi
:
10.3
5741
/
i
s
s
n.025
8
-
27
24.56
.1.28
.
[
27]
A
.
P
.
V
a
ne
ga
s
,
“
P
r
e
di
c
t
i
on
of
phys
i
c
a
l
a
nd
c
he
m
i
c
a
l
pa
r
a
m
e
t
e
r
s
of
w
a
t
e
r
qua
l
i
t
y
us
i
ng
r
e
m
ot
e
s
e
ns
or
s
:
c
a
s
e
s
t
udy
of
t
he
N
e
us
a
r
e
s
e
r
voi
r
,”
M
.Sc
.
T
he
s
i
s
,
M
a
e
s
t
r
í
a
e
n
C
i
e
nc
i
a
s
A
m
bi
e
nt
a
l
e
s
,
F
a
c
ul
t
a
d
de
C
i
e
nc
i
a
s
N
a
t
ur
a
l
e
s
e
I
nge
ni
e
r
í
a
,
J
or
ge
T
a
de
o
L
oz
a
n
o
U
ni
ve
r
s
i
t
y F
ounda
t
i
on of
B
ogot
a
, C
ol
om
bi
a
, 2025.
[
28]
D
.
C
.
R
.
R
a
m
í
r
e
z
,
“
M
e
t
hod
f
or
e
s
t
i
m
a
t
i
ng
t
ot
a
l
s
u
s
pe
nde
d
s
ol
i
ds
a
s
a
n
i
ndi
c
a
t
or
of
w
a
t
e
r
qua
l
i
t
y
us
i
ng
s
a
t
e
l
l
i
t
e
i
m
a
ge
r
y
(
i
n
S
pa
ni
s
h:
M
é
t
odo
de
e
s
t
i
m
ac
i
ón
de
s
ól
i
dos
s
us
pe
ndi
do
s
t
ot
al
e
s
c
om
o
i
ndi
c
ador
de
l
a
c
al
i
dad
de
l
agua
m
e
di
ant
e
i
m
áge
ne
s
s
at
e
l
i
t
al
e
s
)
,”
M
.Sc
.
T
he
s
i
s
,
F
a
c
ul
t
a
d
de
C
i
e
nc
i
a
s
A
gr
a
r
i
a
s
,
E
s
c
ue
l
a
de
pos
gr
a
do,
U
ni
ve
r
s
i
da
d
N
a
c
i
ona
l
de
C
ol
om
bi
a
,
B
ogot
á
,
C
ol
om
bi
a
, 2017.
[
29]
M
. G
hol
i
z
a
de
h,
A
. M
e
l
e
s
s
e
, a
nd
L
.
R
e
ddi
, “
A
c
om
pr
e
he
ns
i
ve
r
e
vi
e
w
on w
a
t
e
r
qua
l
i
t
y pa
r
a
m
e
t
e
r
s
e
s
t
i
m
a
t
i
on u
s
i
ng r
e
m
ot
e
s
e
ns
i
n
g
t
e
c
hni
que
s
,”
Se
n
s
or
s
, vol
. 16, no. 8, A
ug. 2016, doi
:
10.3390/
s
16081298.
[
30]
A
.
K
ul
ka
r
ni
,
“
W
a
t
e
r
qua
l
i
t
y
r
e
t
r
i
e
va
l
f
r
om
L
a
nds
a
t
T
M
i
m
a
ge
r
y,”
P
r
oc
e
di
a
C
om
put
e
r
Sc
i
e
nc
e
,
vol
.
6,
pp.
475
–
480,
2011,
doi
:
10.1016/
j
.pr
oc
s
.2011.08.088.
[
31]
A
.
S
.
Q
a
m
ba
r
a
nd
M
.
M
.
M
.
A
l
K
ha
l
i
dy,
“
D
e
ve
l
opm
e
nt
of
l
oc
a
l
a
nd
gl
oba
l
w
a
s
t
e
w
a
t
e
r
bi
oc
he
m
i
c
a
l
oxyge
n
de
m
a
nd
r
e
a
l
-
t
i
m
e
pr
e
di
c
t
i
on
m
ode
l
s
us
i
ng
s
upe
r
vi
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
s
,”
E
ngi
ne
e
r
i
n
g
A
ppl
i
c
at
i
ons
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
,
vol
.
118
,
F
e
b. 2023, doi
:
10.1016/
j
.e
nga
ppa
i
.2022.105709.
[
32]
F
.
Z
hou,
B
.
L
i
u,
a
nd
K
.
D
ua
n,
“
C
oupl
i
ng
w
a
ve
l
e
t
t
r
a
ns
f
or
m
a
nd
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k
f
or
f
or
e
c
a
s
t
i
ng
e
s
t
ua
r
i
ne
s
a
l
i
ni
t
y,”
J
our
nal
of
H
y
dr
ol
ogy
, vol
. 588, 2020, doi
:
10.1016/
j
.j
hydr
ol
.2020.125127.
[
33]
W
.
H
ua
ng
a
nd
S
.
F
oo,
“
N
e
ur
a
l
ne
t
w
or
k
m
ode
l
i
ng
of
s
a
l
i
ni
t
y
va
r
i
a
t
i
on
i
n
A
p
a
l
a
c
hi
c
ol
a
R
i
ve
r
,”
W
at
e
r
R
e
s
e
a
r
c
h
,
vol
.
36,
no.
1
,
pp. 356
–
362, J
a
n. 2002, doi
:
10.1016/
S
0043
-
1354
(
01)
00195
-
6.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Julio
Cesar
Anaya
-
Valenzuela
is
a
cadastral
Engineer
and
Geodesist
from
the
Universidad
Distrital
Franc
isco
José
de
Caldas,
specia
list
in
engine
er
ing
projec
t
manage
ment
from
the
same
university
and
master
in
remote
sensing
from
the
Universidad
Católica
de
Manizales,
Colombia.
He
works
as
a
public
servant
in
the
govern
or'
s
office
of
Atlántico,
Colombia,
on
issues
associa
ted
with
risk
manage
ment
and
climate
change
.
His
area
of
interes
t
includes re
mote sensing a
pplied
to risk
manageme
nt and
environme
nt,
geographic information
systems
and
ar
tificial
intelligence
algorithms.
He
can
be
contacte
d
at
email:
ju
lio.anaya@ucm.edu.co
.
Gloria
Yaneth
Florez
-
Yepes
development
and
the
environment
,
Ph.D.
Sustaina
ble
Devel
opment
,
Universidad
de
Manizales,
Manizales,
in
2
018.
She
is
professor
at
the
Universidad
Católica
de
Manizales
Colombia,
coordinator
in
the
research
group
on
Technological
and
Environmental
Development.
She
published
mor
e
than
25
scientific
and
research publicat
ions. S
he can be contacted
at email:
gyflorez@
ucm.edu.co.
Yeison
Alberto
Garcés
-
Gómez
received
bachelor’s
degree
in
Electronic
Engineering
,
and
master’s
degrees
and
Ph
.
D
.
in
Engineering
from
E
lectrical,
Electronic
and
Computer
Enginee
ring
Depar
tment,
Universi
d
ad
Naciona
l
de
Colombia,
Manizal
es,
Colombia,
in
2009,
2011
and
2015,
respec
tively.
He
is
full
professor
a
t
the
Academic
Unit
for
Training
in
Natural
Sciences
and
Mathematics,
Universidad
Cató
lica
de
Manizales,
and
teaches
several
courses
such
as
experiment
al
design,
statis
tics
and
ph
ysics
.
His
main
research
focus
is
on
applied
technologies,
embedded
system,
power
electronics
,
power
quality,
but
also
many
other
areas
o
f
electronics,
signal
processing
and
didactics.
He
published
more
than
30
scientific
and
research
publications,
among
them
more
than
10
jo
urnal
papers.
He
worked
as
principal
researcher
on
commercial
projects
and
projects
by
the
M
inistry
of
Science,
Tec
h
and Innov
ation,
Republi
c of Col
ombia.
He can be contacted at email:
ygarces@ucm.edu.co
.
Evaluation Warning : The document was created with Spire.PDF for Python.