I
nte
rna
t
io
na
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
15
,
No
.
1
,
Ma
r
ch
20
26
,
p
p
.
1
4
2
~
1
5
4
I
SS
N:
2252
-
8
8
1
4
,
DOI
:
1
0
.
1
1
5
9
1
/ijaas
.
v15.
i
1
.
pp
142
-
1
5
4
142
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
a
s
.
ia
esco
r
e.
co
m
Ens
emble ma
chin
e learning
bas
ed
mo
del t
o
estima
te
irriga
tion
wa
ter
re
quireme
nt
for
whea
t
crop
Sa
t
endra
K
um
a
r
J
a
in,
Anil K
um
a
r
G
up
t
a
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s,
B
a
r
k
a
t
u
l
l
a
h
U
n
i
v
e
r
si
t
y
,
B
h
o
p
a
l
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
1
5
,
2
0
2
5
R
ev
is
ed
No
v
1
8
,
2
0
2
5
Acc
ep
ted
J
an
1
,
2
0
2
6
In
d
ia
fa
c
e
s
a
se
rio
u
s
wa
ter
sh
o
rta
g
e
issu
e
,
a
s
it
s
p
o
p
u
latio
n
g
r
o
ws
f
a
ste
r
th
a
n
th
e
p
e
rc
e
n
tag
e
o
f
fre
sh
wa
ter
a
v
a
il
a
b
le,
wit
h
o
n
l
y
4
%
o
f
th
e
wo
rld
'
s
fre
sh
wa
ter
a
v
a
il
a
b
le
to
1
8
%
o
f
th
e
wo
rld
'
s
p
o
p
u
lati
o
n
.
A
g
ricu
lt
u
re
se
c
to
r
is
m
o
re
wa
ter
-
c
o
n
su
m
in
g
se
c
to
r
i
n
In
d
ia
.
In
d
ia'
s
irri
g
a
ti
o
n
s
y
ste
m
stil
l
f
a
c
e
s
two
sig
n
ifi
c
a
n
t
p
ro
b
lem
s:
lo
w
irri
g
a
t
io
n
e
fficie
n
c
y
a
n
d
a
lac
k
o
f
o
p
t
imiz
a
ti
o
n
d
u
ri
n
g
irri
g
a
ti
o
n
.
T
o
a
d
d
re
ss
th
e
se
p
ro
b
lem
s
,
a
g
ricu
l
tu
rists
o
u
g
h
t
t
o
b
e
a
wa
re
o
f
th
e
wa
ter
re
q
u
irem
e
n
ts
fo
r
c
ro
p
s
b
e
fo
re
h
a
n
d
.
I
n
n
o
v
a
ti
v
e
fi
e
ld
s
li
k
e
m
a
c
h
in
e
lea
rn
in
g
,
a
b
ra
n
c
h
o
f
a
r
ti
ficia
l
in
tell
ig
e
n
c
e
,
h
a
v
e
a
b
ig
p
o
ten
ti
a
l
to
imp
ro
v
e
irri
g
a
ti
o
n
.
Ve
rify
i
n
g
t
h
e
su
it
a
b
il
it
y
o
f
t
h
e
g
ra
d
ien
t
b
o
o
sti
n
g
re
g
re
ss
o
r
m
a
c
h
in
e
lea
rn
in
g
a
l
g
o
r
it
h
m
-
b
a
se
d
m
o
d
e
l
fo
r
e
stim
a
ti
n
g
irri
g
a
ti
o
n
wa
ter
re
q
u
irem
e
n
ts
(IW
R)
is
th
e
a
im
o
f
th
is
re
se
a
rc
h
.
Th
e
e
x
p
e
rime
n
t
is
c
o
n
d
u
c
ted
i
n
Lu
d
h
ia
n
a
,
a
c
it
y
in
Ce
n
tral
P
u
n
jab
,
I
n
d
ia
,
wit
h
a
h
o
t,
se
m
i
-
a
rid
c
li
m
a
te
th
a
t
fe
a
tu
re
s
sc
o
rc
h
in
g
su
m
m
e
rs
a
n
d
c
h
il
ly
wi
n
ters
.
T
h
e
re
su
lt
s
d
e
m
o
n
stra
te
th
e
re
m
a
rk
a
b
ly
h
ig
h
a
c
c
u
ra
c
y
ra
te
with
c
o
e
fficie
n
t
o
f
d
e
term
in
a
ti
o
n
(
R
2
)
=
0
.
9
8
f
o
r
p
re
d
ictin
g
IW
R
.
T
h
e
su
g
g
e
ste
d
m
o
d
e
l
,
wh
ich
is
b
a
se
d
o
n
a
g
ra
d
ie
n
t
b
o
o
st
in
g
re
g
re
ss
io
n
,
a
ll
o
ws
th
e
sta
k
e
h
o
ld
e
rs
t
o
a
c
c
u
ra
tely
e
stim
a
te
th
e
a
m
o
u
n
t
o
f
wa
ter
n
e
e
d
e
d
fo
r
irri
g
a
ti
o
n
,
t
h
e
n
u
m
b
e
r
o
f
irri
g
a
ti
o
n
a
p
p
li
c
a
ti
o
n
s
f
o
r
th
e
g
ro
wi
n
g
se
a
so
n
o
f
wh
e
a
t
c
ro
p
s
,
a
n
d
th
e
in
terv
a
l
b
e
twe
e
n
irri
g
a
ti
o
n
s.
K
ey
w
o
r
d
s
:
C
r
o
p
co
ef
f
icien
t
Gr
ad
ien
t b
o
o
s
tin
g
r
eg
r
ess
o
r
I
r
r
ig
atio
n
wate
r
r
eq
u
i
r
em
en
t
Ma
ch
in
e
lear
n
in
g
R
ef
er
en
ce
ev
ap
o
t
r
an
s
p
ir
atio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Saten
d
r
a
Ku
m
ar
J
ain
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
Ap
p
licatio
n
s
,
B
ar
k
atu
llah
Un
iv
er
s
ity
B
h
o
p
al,
Ma
d
h
y
a
Pra
d
esh
,
I
n
d
i
a
E
m
ail:
s
aten
d
r
a.
k
.
jain
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
d
ia
is
in
a
u
n
iq
u
e
p
o
s
itio
n
to
p
r
o
d
u
ce
wh
ea
t
g
lo
b
ally
.
I
t
is
g
r
o
wn
in
m
o
s
t
n
o
r
th
er
n
I
n
d
ian
s
tates,
in
clu
d
in
g
Pu
n
jab
,
Har
y
a
n
a,
M
ad
h
y
a
Pra
d
esh
,
a
n
d
R
ajasth
an
.
W
h
ea
t
is
a
r
a
b
i
cr
o
p
w
h
ich
i
s
s
o
wn
in
Octo
b
er
an
d
m
at
u
r
e
s
in
Ma
r
ch
.
I
t
is
b
eliev
ed
t
h
at
wh
ea
t
r
eq
u
ir
es
a
s
u
f
f
icien
t
am
o
u
n
t
o
f
wate
r
to
g
r
o
w.
I
n
I
n
d
ia,
it
h
as
b
ee
n
o
b
s
er
v
ed
th
at
f
ar
m
e
r
s
ar
e
g
en
er
ally
u
n
awa
r
e
o
f
th
e
am
o
u
n
t
o
f
wate
r
t
h
at
is
s
p
r
ay
ed
o
n
cr
o
p
s
d
u
r
in
g
ir
r
ig
atio
n
.
Sig
n
if
ican
t
wate
r
l
o
s
s
d
u
r
in
g
ir
r
ig
atio
n
h
as
also
b
ee
n
n
o
ted
as
a
r
esu
lt
o
f
s
p
ills
,
leak
s
,
an
d
ev
ap
o
r
atio
n
.
L
ess
ir
r
ig
atio
n
ef
f
icien
cy
r
esu
lts
f
r
o
m
it.
T
h
ese
p
r
o
b
le
m
s
in
s
p
ir
e
d
u
s
to
s
u
g
g
est
a
m
o
d
el
b
ased
o
n
ar
tific
ial
in
tellig
en
ce
t
h
at
g
iv
es
u
s
r
em
ar
k
ab
le
p
r
ec
is
io
n
r
e
g
ar
d
in
g
th
e
am
o
u
n
t
o
f
wate
r
u
s
ed
d
u
r
in
g
ir
r
ig
atio
n
.
C
alcu
latin
g
th
e
f
r
e
q
u
en
cy
o
f
ir
r
ig
atio
n
ap
p
licati
o
n
s
an
d
th
e
tim
e
b
etwe
en
ir
r
i
g
atio
n
s
m
ay
also
b
e
ma
d
e
ea
s
ier
b
y
t
h
e
m
o
d
el.
T
o
ac
cu
r
ately
d
eter
m
in
e
th
e
am
o
u
n
t
o
f
wate
r
n
ee
d
e
d
f
o
r
ir
r
ig
atio
n
,
a
n
u
m
b
er
o
f
m
eteo
r
o
lo
g
ical
v
a
r
iab
les
m
u
s
t
b
e
co
n
s
id
er
ed
,
i
n
clu
d
in
g
tem
p
er
atu
r
e,
win
d
s
p
ee
d
,
h
u
m
id
it
y
,
s
u
n
s
h
in
e
h
o
u
r
s
,
p
r
ec
ip
itatio
n
,
an
d
r
e
f
er
en
ce
e
v
ap
o
tr
an
s
p
ir
atio
n
(
E
T
o
)
.
T
h
e
liter
atu
r
e
co
n
tain
s
a
n
u
m
b
er
o
f
d
ir
ec
t
an
d
in
d
ir
ec
t
tech
n
iq
u
es f
o
r
ca
lcu
latin
g
E
T
o
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
E
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
mo
d
el
to
esti
ma
te
ir
r
ig
a
tio
n
w
a
ter r
eq
u
ir
eme
n
t
… (
S
a
te
n
d
r
a
K
u
ma
r
Ja
in
)
143
A
co
m
m
o
n
ly
u
s
ed
tech
n
iq
u
e
t
h
at
d
ep
en
d
s
o
n
clim
atic
p
a
r
a
m
eter
s
is
th
e
Pen
m
an
-
Mo
n
teit
h
eq
u
atio
n
,
Allen
et
a
l.
[
1
]
s
u
g
g
ested
b
y
th
e
Fo
o
d
an
d
Ag
r
icu
ltu
r
e
O
r
g
an
izatio
n
(
FAO)
,
Un
ited
Natio
n
s
.
T
h
e
wate
r
r
eq
u
ir
em
e
n
t
o
f
a
cr
o
p
,
e
x
p
r
e
s
s
ed
as
cr
o
p
ev
ap
o
tr
an
s
p
ir
ati
o
n
(
ETc
)
,
is
ca
lcu
lated
b
y
m
u
ltip
ly
in
g
its
c
r
o
p
co
ef
f
icien
t
(
Kc
)
b
y
E
T
o
.
T
o
g
eth
er
with
ef
f
ec
tiv
e
r
ain
f
all,
E
T
c
is
a
cr
u
cial
p
ar
am
eter
f
o
r
d
eter
m
in
i
n
g
th
e
am
o
u
n
t
o
f
wate
r
n
ee
d
ed
f
o
r
ir
r
ig
atio
n
.
T
h
e
r
e
h
a
v
e
b
e
en
n
u
m
e
r
o
u
s
a
u
th
o
r
s
w
h
o
h
a
v
e
c
o
n
t
r
ib
u
t
e
d
s
i
g
n
i
f
i
ca
n
t
ly
to
t
h
e
e
s
t
im
a
t
i
o
n
o
f
i
r
r
i
g
a
t
i
o
n
w
a
t
er
r
eq
u
i
r
e
m
en
ts
(
I
W
R
)
th
r
o
u
g
h
t
h
e
u
s
e
o
f
d
i
r
e
c
t,
in
d
ir
e
c
t
,
an
d
m
a
ch
i
n
e
l
ea
r
n
i
n
g
-
b
a
s
e
d
m
e
t
h
o
d
s
,
a
c
co
r
d
i
n
g
to
th
e
l
i
te
r
a
t
u
r
e
r
ev
i
e
w
.
S
a
g
g
i
an
d
J
ai
n
[
2
]
c
o
m
p
le
t
e
d
a
t
h
o
r
o
u
g
h
i
n
v
e
s
t
i
g
a
t
io
n
i
n
to
s
m
a
r
t
i
r
r
i
g
a
t
io
n
s
ch
e
d
u
l
i
n
g
d
e
c
i
s
i
o
n
s
u
p
p
o
r
t
s
y
s
t
em
s
.
S
a
tp
u
t
e
e
t
a
l
.
[
3
]
e
v
a
lu
a
t
ed
t
h
e
n
e
t
a
n
d
g
r
o
s
s
w
a
t
e
r
i
r
r
i
g
a
t
i
o
n
r
eq
u
ir
e
m
en
t
s
f
o
r
s
i
x
t
e
e
n
d
i
s
t
i
n
c
t
cr
o
p
s
i
n
C
en
t
r
a
l
P
u
n
j
ab
,
b
o
t
h
w
i
th
an
d
w
i
t
h
o
u
t
e
f
f
e
c
t
i
v
e
r
a
i
n
f
a
l
l
.
A
k
b
a
r
e
t
a
l.
[
4
]
u
s
e
d
C
r
o
p
W
a
t
s
o
f
t
wa
r
e
t
o
e
s
ti
m
a
t
e
c
r
o
p
w
a
t
e
r
r
e
q
u
i
r
em
en
t
s
an
d
i
r
r
i
g
a
t
io
n
s
c
h
e
d
u
l
i
n
g
f
o
r
wh
e
a
t
in
P
u
n
j
ab
S
t
a
t
e
,
P
ak
i
s
t
a
n
,
a
cr
o
s
s
v
ar
i
o
u
s
c
l
i
m
a
t
i
c
z
o
n
e
s
.
F
ar
u
q
e
t
a
l.
[
5
]
e
s
t
i
m
at
e
d
th
e
c
r
o
p
w
a
t
er
r
e
q
u
i
r
e
m
en
t
f
o
r
r
i
c
e
i
n
B
a
n
d
a
r
b
a
n
,
B
an
g
l
ad
e
s
h
,
u
s
i
n
g
th
e
s
am
e
s
o
f
t
w
ar
e
.
S
o
l
a
n
g
i
e
t
a
l.
[
6
]
c
a
l
c
u
l
a
t
ed
th
e
am
o
u
n
t
o
f
w
a
te
r
n
e
e
d
ed
f
o
r
t
h
e
i
r
r
i
g
a
t
io
n
o
f
w
h
e
a
t,
co
t
t
o
n
,
b
a
n
a
n
a
s
,
a
n
d
s
u
g
ar
c
a
n
e
.
Sh
ar
m
a
et
a
l.
[
7
]
p
r
o
p
o
s
ed
v
ar
y
in
g
Kc
v
alu
es
f
o
r
ev
e
r
y
s
tag
e
o
f
wh
ea
t
cr
o
p
g
r
o
wth
in
J
alan
d
h
ar
,
Pu
n
jab
,
u
n
d
er
d
r
ip
i
r
r
ig
atio
n
.
A
lo
t
o
f
r
esear
c
h
h
as
b
ee
n
d
o
n
e
in
[
8
]
to
s
ee
h
o
w
th
e
E
T
o
,
m
o
is
tu
r
e
in
d
e
x
,
an
d
ar
id
ity
in
d
e
x
c
h
an
g
e
in
Pu
n
ja
b
S
tate'
s
v
ar
io
u
s
ag
r
o
clim
atic
zo
n
es.
I
n
o
r
d
e
r
to
d
eter
m
i
n
e
th
e
o
p
tim
al
o
p
tio
n
f
o
r
esti
m
atin
g
E
T
o
,
I
s
lam
an
d
Alam
[
9
]
e
v
alu
ate
d
th
e
p
er
f
o
r
m
an
ce
o
f
f
if
teen
em
p
ir
ical
m
o
d
els.
I
r
m
ak
et
a
l.
[
1
0
]
s
u
g
g
ested
s
o
lar
an
d
n
et
r
ad
iatio
n
-
b
ased
eq
u
atio
n
s
to
esti
m
ate
E
T
o
.
Sag
g
i
an
d
J
a
in
[
1
1
]
p
r
o
p
o
s
ed
a
f
u
zz
y
-
g
en
etic
,
r
eg
u
lar
izatio
n
,
an
d
r
an
d
o
m
f
o
r
est
-
b
ased
m
o
d
el
to
esti
m
ate
E
T
c
f
o
r
m
aiz
e
an
d
wh
ea
t
cr
o
p
s
.
Sag
g
i
an
d
J
ain
[
1
2
]
p
r
esen
te
d
d
ee
p
lear
n
in
g
b
ased
m
o
d
e
l
to
esti
m
ate
E
T
o
.
Pag
an
o
e
t
a
l.
[
1
3
]
u
tili
ze
d
m
u
lti
-
lay
er
p
er
ce
p
tr
o
n
a
n
d
r
an
d
o
m
f
o
r
est
t
o
f
o
r
ec
ast
d
ai
ly
ac
tu
al
ev
ap
o
tr
a
n
s
p
ir
atio
n
.
Me
en
al
et
a
l.
[
1
4
]
d
is
cu
s
s
ed
th
e
ap
p
licab
ilit
y
o
f
r
an
d
o
m
f
o
r
est
f
o
r
ev
a
p
o
tr
an
s
p
ir
atio
n
m
o
d
ellin
g
.
R
av
in
d
r
a
n
et
a
l.
[
1
5
]
t
ested
d
ee
p
n
eu
r
al
n
etwo
r
k
-
b
ased
m
o
d
el
b
u
il
t
o
n
d
if
f
e
r
en
t
clim
atic
p
ar
am
eter
s
to
p
r
e
d
ict
E
T
o
.
Ku
m
ar
et
a
l.
[
1
6
]
a
p
p
lied
an
ar
tific
ial
n
eu
r
al
n
e
tw
o
r
k
f
o
r
esti
m
atio
n
o
f
E
T
o
o
f
th
e
wester
n
Him
alay
an
r
eg
io
n
.
Gr
an
ata
[
1
7
]
c
r
ea
ted
th
r
ee
m
o
d
els b
ased
o
n
v
ar
i
o
u
s
clim
ate
p
ar
am
eter
s
an
d
u
s
ed
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
e
s
,
r
eg
r
ess
io
n
tr
ee
s
,
b
ag
g
in
g
,
an
d
r
an
d
o
m
f
o
r
ests
to
esti
m
ate
E
T
o
.
Kh
o
s
r
av
i
et
a
l.
[
1
8
]
e
v
alu
ated
th
e
ca
p
ac
ity
o
f
a
d
ap
tiv
e
n
eu
r
o
-
f
u
zz
y
in
f
er
en
ce
s
y
s
tem
s
an
d
d
if
f
er
en
t
d
ata
m
in
i
n
g
alg
o
r
ith
m
s
to
esti
m
ate
E
T
o
.
F
er
r
eir
a
et
a
l.
[
1
9
]
e
s
tim
ated
E
T
o
i
n
B
r
az
il
u
s
i
n
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
an
d
a
r
tific
ial
in
tellig
en
ce
.
Hu
et
a
l.
[
2
0
]
h
a
ve
co
n
tr
asted
esti
m
ated
E
T
o
with
p
h
y
s
ical
-
b
ased
,
d
ata
-
d
r
iv
e
n
-
b
ased
,
an
d
h
y
b
r
id
m
o
d
elin
g
.
To
p
r
ed
ict
E
T
o
,
m
o
d
els b
ased
o
n
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
,
r
an
d
o
m
f
o
r
ests
,
lass
o
r
id
g
e,
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
s
wer
e
u
s
ed
in
[
2
1
]
.
Ku
m
a
r
et
a
l.
[
2
2
]
e
s
tim
ated
E
T
o
u
s
in
g
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es,
r
an
d
o
m
f
o
r
ests
,
g
r
a
d
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
s
,
an
d
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
n
eu
r
al
n
etwo
r
k
s
.
Hu
et
a
l.
[
2
3
]
s
u
g
g
ested
an
I
o
T
-
b
ased
m
o
d
e
l
th
at
esti
m
ates
E
T
o
u
s
in
g
a
m
ac
h
in
e
-
lear
n
i
n
g
alg
o
r
ith
m
.
Nag
ap
p
an
et
a
l.
[
2
4
]
i
n
v
esti
g
ate
h
o
w
well
d
ee
p
n
eu
r
al
n
etwo
r
k
s
wo
r
k
f
o
r
s
ch
ed
u
lin
g
ir
r
ig
atio
n
.
B
astam
et
a
l.
[
2
5
]
p
r
o
p
o
s
ed
a
h
y
b
r
id
en
s
em
b
le
f
r
am
ewo
r
k
f
o
r
s
m
ar
t ir
r
ig
atio
n
.
T
h
is
s
tu
d
y
'
s
g
o
al
is
to
cr
ea
te
an
ar
tific
ial
in
tellig
en
ce
-
b
ased
m
o
d
el
f
o
r
esti
m
atin
g
ir
r
i
g
at
io
n
wate
r
ne
ed
s
to
s
av
e
f
r
esh
wate
r
f
o
r
th
e
n
ex
t
g
en
er
atio
n
.
As
ag
r
icu
l
tu
r
e
s
ec
to
r
is
ex
tr
ac
tin
g
h
u
g
e
u
n
d
er
g
r
o
u
n
d
wate
r
f
o
r
ir
r
ig
atio
n
,
we
wan
t
to
r
ed
u
ce
it
b
y
p
r
ec
is
e
p
r
e
d
ictio
n
u
s
i
n
g
p
r
o
p
o
s
ed
m
o
d
el.
T
h
e
r
em
a
in
in
g
p
o
r
tio
n
o
f
th
e
ar
ticle
is
ex
p
lain
ed
in
th
e
f
o
llo
win
g
o
r
d
e
r
:
th
e
d
ataset
o
f
th
e
s
ite
u
n
d
er
s
tu
d
y
,
th
e
m
ac
h
in
e
lear
n
in
g
-
b
ased
m
o
d
el
th
at
is
s
u
g
g
ested
t
o
esti
m
ate
th
e
am
o
u
n
t
o
f
wate
r
n
ee
d
ed
f
o
r
ir
r
ig
atio
n
,
th
e
g
r
a
d
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
alg
o
r
ith
m
,
th
e
r
esu
lts
th
at
ar
e
d
is
cu
s
s
ed
,
an
d
th
e
s
tu
d
y
'
s
co
n
clu
s
io
n
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
Pu
n
jab
,
a
s
tate
in
n
o
r
th
er
n
I
n
d
ia,
is
well
k
n
o
wn
f
o
r
its
co
n
tr
ib
u
tio
n
to
th
e
g
r
ee
n
r
ev
o
lu
tio
n
.
I
t
s
ig
n
if
ican
tly
b
o
o
s
ts
th
e
n
atio
n
'
s
p
r
o
d
u
ctio
n
o
f
r
ice
a
n
d
wh
e
at.
A
s
u
f
f
icien
t
am
o
u
n
t
o
f
wa
ter
is
n
ec
ess
ar
y
f
o
r
th
e
g
r
o
wth
o
f
ev
er
y
c
r
o
p
.
N
u
m
er
o
u
s
clim
atic
p
ar
a
m
eter
s
,
c
r
o
p
ch
ar
ac
te
r
is
tics
,
an
d
g
eo
g
r
ap
h
ic
d
ata
ab
o
u
t
th
e
s
tu
d
y
s
ite
ar
e
n
ee
d
ed
to
co
m
p
lete
th
is
task
.
T
h
e
d
is
tr
ict
ch
o
s
en
f
o
r
t
h
is
s
tu
d
y
is
L
u
d
h
ian
a
in
th
e
Pu
n
ja
b
s
tate,
wh
ich
h
as
h
o
t,
s
em
i
-
ar
id
a
g
r
o
clim
atic
zo
n
e.
I
ts
av
er
a
g
e
e
lev
atio
n
is
2
4
0
m
eter
s
,
a
n
d
i
ts
co
o
r
d
in
ates
ar
e
30
°
5
4
'
N
7
5
°
5
1
'
E
.
T
h
e
av
er
a
g
e
am
o
u
n
t
o
f
p
r
e
cip
itatio
n
o
b
s
er
v
ed
ea
ch
y
ea
r
is
5
0
0
-
7
5
0
m
m
.
Fro
m
m
id
-
Ma
r
c
h
th
r
o
u
g
h
th
e
last
wee
k
o
f
J
u
n
e,
th
e
h
ea
t
wav
es
ar
r
i
v
e.
T
h
e
s
o
u
th
-
west
m
o
n
s
o
o
n
p
r
o
v
id
es
7
8
%
o
f
th
e
p
r
ec
ip
itatio
n
th
at
f
alls
b
etwe
en
J
u
ly
an
d
Sep
tem
b
e
r
.
Du
r
in
g
th
e
n
o
n
-
m
o
n
s
o
o
n
m
o
n
th
s
,
w
ester
n
d
is
tu
r
b
an
ce
s
ac
co
u
n
t f
o
r
th
e
r
em
ain
in
g
2
2
% o
f
p
r
ec
ip
itatio
n
.
A
d
ataset
s
p
an
n
in
g
twen
ty
-
tw
o
y
ea
r
s
,
f
r
o
m
2
0
0
0
to
2
0
2
1
,
h
as
b
ee
n
co
n
s
id
er
e
d
f
o
r
ex
p
e
r
im
en
tatio
n
.
T
h
e
d
ataset
is
f
u
r
n
is
h
ed
with
s
ix
p
a
r
am
et
er
s
:
tem
p
er
at
u
r
e
(
b
o
t
h
lo
w
a
n
d
h
ig
h
)
,
h
u
m
id
ity
,
win
d
s
p
ee
d
,
s
u
n
s
h
in
e
h
o
u
r
s
,
an
d
p
r
ec
ip
ita
tio
n
.
An
n
u
al
m
ea
n
v
alu
es
o
f
th
ese
p
ar
am
eter
s
ar
e
s
h
o
wn
in
T
ab
le
1
,
an
d
th
e
m
ea
n
m
o
n
th
ly
v
a
r
iatio
n
is
d
ep
icted
in
Fig
u
r
e
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J
Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
142
-
1
5
4
144
T
ab
le
1
.
L
u
d
h
ian
a'
s
an
n
u
al
m
e
an
clim
ate
d
at
a
v
alu
es
Y
e
a
r
M
a
x
.
t
e
m
p
e
r
a
t
u
r
e
(
°
C
)
M
i
n
.
t
e
mp
e
r
a
t
u
r
e
(
°
C
)
R
e
l
a
t
i
v
e
h
u
mi
d
i
t
y
(
%)
W
i
n
d
(
m
/
s)
ETo
(
mm
/
d
a
y
)
2
0
0
0
3
4
.
3
8
1
9
.
3
0
3
3
.
2
0
2
.
0
6
6
.
1
8
2
0
0
1
3
3
.
9
9
1
9
.
3
2
3
4
.
7
4
1
.
8
7
5
.
9
2
2
0
0
2
3
4
.
9
0
2
0
.
0
7
3
3
.
1
2
1
.
9
1
6
.
0
9
2
0
0
3
3
3
.
0
8
1
9
.
0
5
4
1
.
3
9
1
.
9
6
5
.
7
0
2
0
0
4
3
4
.
0
7
1
9
.
7
1
3
7
.
5
4
1
.
9
3
5
.
9
1
2
0
0
5
3
2
.
6
7
1
8
.
9
0
4
1
.
6
4
1
.
9
0
5
.
6
4
2
0
0
6
3
2
.
9
8
1
9
.
6
7
4
4
.
6
7
1
.
9
2
5
.
6
2
2
0
0
7
3
3
.
3
1
1
9
.
5
6
4
1
.
7
8
1
.
8
9
5
.
7
0
2
0
0
8
3
1
.
8
0
1
8
.
5
8
4
7
.
8
1
1
.
8
6
5
.
3
5
2
0
0
9
3
3
.
6
9
1
9
.
5
4
3
8
.
2
3
1
.
8
7
5
.
8
3
2
0
1
0
3
3
.
3
7
1
9
.
6
7
4
1
.
9
7
1
.
8
6
5
.
7
1
2
0
1
1
3
2
.
0
6
1
8
.
9
2
4
6
.
7
7
1
.
7
8
5
.
3
7
2
0
1
2
3
2
.
8
2
1
8
.
7
5
3
9
.
9
7
2
.
0
8
5
.
8
7
2
0
1
3
3
2
.
2
6
1
8
.
6
5
4
7
.
1
1
1
.
9
8
5
.
5
9
2
0
1
4
3
2
.
2
7
1
8
.
5
2
4
4
.
5
4
2
.
0
0
5
.
7
0
2
0
1
5
3
2
.
0
0
1
8
.
4
9
4
7
.
7
1
1
.
9
4
5
.
4
4
2
0
1
6
3
3
.
6
9
1
9
.
5
0
3
9
.
6
2
1
.
9
3
5
.
8
9
2
0
1
7
3
2
.
7
0
1
8
.
9
6
4
4
.
3
6
1
.
9
0
5
.
6
4
2
0
1
8
3
2
.
0
7
1
8
.
6
7
4
7
.
0
3
1
.
9
2
5
.
5
5
2
0
1
9
3
0
.
8
1
1
8
.
3
7
5
3
.
9
3
1
.
8
2
5
.
1
8
2
0
2
0
3
0
.
8
7
1
8
.
1
9
5
4
.
5
1
1
.
9
2
5
.
1
4
2
0
2
1
3
2
.
1
0
1
8
.
7
6
4
9
.
9
9
1
.
9
1
5
.
4
6
Fig
u
r
e
1
.
Av
e
r
ag
e
m
o
n
th
ly
ch
an
g
es in
L
u
d
h
ian
a'
s
E
T
o
,
h
u
m
id
ity
,
tem
p
er
atu
r
e,
an
d
win
d
s
p
ee
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
E
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
mo
d
el
to
esti
ma
te
ir
r
ig
a
tio
n
w
a
ter r
eq
u
ir
eme
n
t
… (
S
a
te
n
d
r
a
K
u
ma
r
Ja
in
)
145
In
(
1
)
is
u
s
ed
to
ca
lcu
late
E
T
o
v
alu
es.
FAO
r
ec
o
m
m
en
d
ed
E
T
o
v
alu
es b
ased
o
n
v
ar
io
u
s
ag
r
o
clim
atic
zo
n
es.
W
h
en
tem
p
er
atu
r
es
a
r
e
h
ig
h
e
r
th
a
n
3
0
d
e
g
r
ee
s
,
it
v
a
r
ies
b
etwe
en
6
an
d
8
m
illi
m
eter
s
p
er
d
ay
i
n
ar
i
d
an
d
s
em
i
-
ar
i
d
r
eg
io
n
s
an
d
5
to
7
m
m
p
e
r
d
a
y
in
th
e
h
u
m
id
r
e
g
io
n
.
T
h
is
r
a
n
g
e
is
co
n
f
ir
m
ed
b
y
T
a
b
le
1
'
s
m
ea
n
E
T
o
v
alu
es.
Pu
n
jab
is
a
m
aj
o
r
wh
ea
t
p
r
o
d
u
ce
r
in
I
n
d
ia.
Fo
r
m
an
y
p
eo
p
le
in
I
n
d
ia,
w
h
ea
t
is
a
s
tap
le
d
iet.
Mo
r
eo
v
er
,
it
is
r
eg
ar
d
ed
as
a
cr
o
p
th
at
ca
n
b
e
u
s
ed
to
p
r
o
d
u
ce
a
wid
e
r
an
g
e
o
f
f
o
o
d
item
s
.
Ma
n
y
ty
p
es
o
f
wh
ea
t
v
ar
iety
,
in
clu
d
in
g
PB
W
8
2
6
,
PB
W
8
6
9
,
PB
W
8
2
4
,
P
B
W
8
0
3
,
an
d
o
th
er
s
,
ar
e
p
r
o
d
u
ce
d
in
th
is
r
eg
io
n
.
T
o
g
r
o
w
wh
ea
t,
a
lar
g
e
am
o
u
n
t
o
f
lan
d
is
n
ee
d
ed
.
As
a
r
esu
lt,
ir
r
ig
atin
g
th
e
c
r
o
p
lan
d
r
e
q
u
ir
es
a
s
ig
n
if
ican
t
am
o
u
n
t
o
f
f
r
esh
wate
r
.
I
t
en
c
o
u
r
ag
es
u
s
f
u
r
th
er
t
o
cr
ea
te
a
m
ac
h
in
e
lear
n
in
g
-
b
ased
m
o
d
e
l
th
at
p
r
ed
icts
h
o
w
m
u
ch
wate
r
cr
o
p
s
will
n
ee
d
to
g
r
o
w.
T
h
e
am
o
u
n
t
o
f
wate
r
n
ee
d
ed
f
o
r
ir
r
ig
atio
n
f
o
r
th
e
wh
ea
t
cr
o
p
is
co
n
s
id
er
ed
in
th
is
s
tu
d
y
.
T
h
e
s
tan
d
ar
d
tim
ef
r
am
e
f
o
r
g
r
o
win
g
an
d
h
ar
v
esti
n
g
wh
ea
t
is
f
r
o
m
th
e
last
wee
k
o
f
Octo
b
er
to
th
e
last
wee
k
o
f
M
ar
ch
.
I
t
tak
es
r
o
u
g
h
ly
1
4
5
to
1
5
0
d
ay
s
in
to
tal.
T
ab
le
2
co
n
ta
in
s
th
e
in
f
o
r
m
atio
n
o
n
wh
ea
t,
wh
ich
is
u
s
ed
in
th
i
s
s
tu
d
y
at
th
e
tim
e
o
f
esti
m
atin
g
cr
o
p
wate
r
r
eq
u
ir
em
en
t.
0
=
0
.
408
×
(
−
)
+
×
(
900
+
273
)
×
2
×
(
−
)
∆
+
×
(
1
+
.
34
×
2
)
(
1
)
T
ab
le
2
.
W
h
ea
t c
r
o
p
d
ata
C
r
o
p
D
u
r
a
t
i
o
n
D
a
t
e
o
f
s
o
w
i
n
g
D
a
t
e
o
f
h
a
r
v
e
s
t
i
n
g
S
o
i
l
t
y
p
e
G
r
o
w
i
n
g
s
t
a
g
e
s
Le
n
g
t
h
o
f
g
r
o
w
t
h
i
n
e
a
c
h
s
t
a
g
e
W
h
e
a
t
1
4
6
d
a
y
s
28
-
O
c
t
21
-
2
2
M
a
r
c
h
S
a
n
d
y
l
o
a
m
I
n
t
i
mal
st
a
g
e
2
5
d
a
y
s
D
e
v
e
l
o
p
i
n
g
s
t
a
g
e
4
1
d
a
y
s
M
i
d
-
se
a
so
n
5
1
d
a
y
s
La
t
e
-
sea
s
o
n
2
9
d
a
y
s
2
.
2
.
P
r
o
po
s
ed
m
o
del
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
is
p
r
esen
ted
in
Fig
u
r
e
2
.
I
t
is
d
iv
id
ed
i
n
to
f
o
u
r
p
h
ases
.
T
h
e
f
ir
s
t
s
tag
e
in
v
o
lv
es
g
ath
er
in
g
th
e
n
ec
ess
ar
y
d
ata.
T
h
e
FAO
co
n
ce
p
t
is
u
s
ed
in
t
h
e
s
ec
o
n
d
s
tag
e
to
ca
lcu
late
o
b
s
er
v
ed
E
T
o
,
E
T
c
(
cr
o
p
wate
r
r
eq
u
ir
em
en
t)
,
an
d
I
W
R
.
T
h
e
th
ir
d
s
tep
s
ep
a
r
ates
th
e
co
m
b
in
e
d
d
ata
(
clim
ate
d
ata
alo
n
g
with
o
b
s
er
v
ed
E
T
o
r
ec
eiv
ed
f
r
o
m
s
ec
o
n
d
s
tag
e)
in
to
two
g
r
o
u
p
s
:
tr
ain
in
g
(
8
0
%)
an
d
test
(
2
0
%)
d
atasets
.
T
h
e
tr
ain
ed
g
r
ad
ien
t
b
o
o
s
tin
g
alg
o
r
ith
m
an
ticip
ates
th
e
n
ew
E
T
o
,
E
T
c,
an
d
I
W
R
v
alu
es
in
th
is
s
tag
e.
At
th
e
f
o
u
r
th
s
tag
e,
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
is
ev
alu
ated
b
ased
o
n
o
b
s
er
v
ed
I
W
R
(
o
b
tain
e
d
f
r
o
m
s
ec
o
n
d
s
tag
e)
an
d
p
r
ed
icted
I
W
R
(
o
b
tain
e
d
f
r
o
m
th
ir
d
s
tag
e)
.
Fig
u
r
e
2
.
E
n
s
em
b
le
m
ac
h
in
e
lear
n
in
g
b
ased
p
r
o
p
o
s
ed
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J
Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
142
-
1
5
4
146
T
o
ca
lcu
late
th
e
am
o
u
n
t
o
f
w
ater
n
ee
d
ed
f
o
r
ir
r
ig
atio
n
,
E
T
o
is
co
m
p
u
ted
f
ir
s
t.
I
t
is
a
p
r
o
ce
s
s
wh
er
e
wate
r
is
lo
s
t
f
r
o
m
th
e
s
o
il
an
d
p
lan
ts
at
th
e
s
am
e
tim
e.
G
eo
g
r
ap
h
ical
in
f
o
r
m
atio
n
(
latitu
d
e,
lo
n
g
itu
d
e,
an
d
altitu
d
e)
an
d
th
e
m
eteo
r
o
lo
g
i
ca
l
in
f
o
r
m
atio
n
(
t
em
p
er
atu
r
e,
h
u
m
id
ity
,
win
d
s
p
ee
d
,
an
d
s
u
n
s
h
in
e
h
o
u
r
s
)
ar
e
n
ee
d
ed
wh
en
esti
m
atin
g
it
f
o
r
a
g
iv
en
ar
ea
.
T
h
e
E
T
o
o
f
L
u
d
h
ian
a
is
d
eter
m
in
ed
in
t
h
is
s
tu
d
y
u
s
in
g
th
e
MS
.
E
x
ce
l
ap
p
licatio
n
,
wh
ich
in
clu
d
es
all
n
ec
ess
ar
y
f
o
r
m
u
la
s
as
r
ec
o
m
m
en
d
ed
f
o
r
th
e
FA
O
-
5
6
PM
ap
p
r
o
ac
h
s
h
o
wn
in
(
1
)
an
d
v
alid
ated
b
y
th
e
FAO
-
r
ec
o
m
m
en
d
ed
C
r
o
p
W
at
8
.
0
s
o
f
twar
e.
E
T
c
v
alu
es
ar
e
d
e
r
iv
e
d
u
s
in
g
(
2
)
,
wh
er
e
Kc
is
th
e
cr
o
p
c
o
ef
f
icien
t,
wh
ich
co
m
p
r
is
es
th
e
cr
o
p
’
s
ch
ar
ac
te
r
is
tics
.
T
h
e
cr
o
p
d
u
r
atio
n
is
d
iv
i
d
ed
in
to
f
o
u
r
s
tag
es
,
n
am
el
y
in
itia
l
s
ta
g
e,
d
e
v
elo
p
m
e
n
t
s
tag
e,
m
i
d
-
s
ea
s
o
n
s
tag
e
,
an
d
late
s
ea
s
o
n
s
tag
e.
E
ac
h
s
tag
e
h
as
o
wn
s
p
ec
if
ic
d
u
r
atio
n
a
n
d
Kc
v
alu
es.
FAO
s
u
g
g
ested
th
e
s
tan
d
ar
d
Kc
v
al
u
es
f
o
r
d
if
f
er
e
n
t
cr
o
p
s
.
W
h
ea
t
is
tak
en
in
to
co
n
s
id
er
atio
n
in
th
is
s
tu
d
y
with
th
e
ch
ar
ac
ter
is
tics
lis
ted
in
T
ab
le
2
.
T
h
e
Kc
v
alu
es
ar
e
in
h
er
ited
f
r
o
m
[
2
6
]
in
th
is
s
tu
d
y
.
E
f
f
ec
t
iv
e
r
ain
f
all
p
lay
s
cr
u
cial
r
o
le
in
ca
lcu
latin
g
I
W
R
.
A
f
r
ac
tio
n
o
f
r
ain
f
all
th
at
is
n
ee
d
ed
to
g
r
o
w
th
e
cr
o
p
a
n
d
r
etain
ed
in
th
e
s
o
il
is
ca
lled
ef
f
ec
tiv
e
r
ain
f
all.
T
h
e
Un
ited
State
s
Dep
ar
tm
en
t
o
f
Ag
r
icu
ltu
r
e
(
USDA)
s
o
il
co
n
s
er
v
atio
n
s
er
v
ice
is
u
s
ed
t
o
ca
l
cu
late
th
e
ef
f
ec
tiv
e
r
ai
n
f
all
g
i
v
en
in
(
3
)
an
d
(
4
).
T
h
is
s
tu
d
y
ca
lcu
lates
it
ev
er
y
ten
d
ay
s
s
in
ce
th
e
cr
o
p
'
s
d
u
r
atio
n
b
eg
a
n
,
an
d
ev
e
r
y
m
o
n
th
also
.
I
f
th
e
ETc
is
less
th
an
th
e
ef
f
ec
tiv
e
r
ain
f
all,
n
o
i
r
r
ig
atio
n
wate
r
is
n
ee
d
ed
.
On
th
e
o
t
h
er
h
an
d
,
I
W
R
is
eq
u
iv
alen
t
to
t
h
e
cr
o
p
wate
r
r
eq
u
ir
em
e
n
t w
h
en
th
e
r
e
is
n
o
ef
f
ec
tiv
e
r
ai
n
f
all.
=
×
(
2
)
=
×
125
−
0
.
2
×
125
<
250
ℎ
(
3
)
=
125
+
0
.
1
×
>
>
250
ℎ
(
4
)
I
n
ad
d
i
ti
o
n
to
t
h
e
tr
ad
iti
o
n
a
l
ap
p
r
o
ac
h
,
th
is
s
t
u
d
y
m
ea
s
u
r
es
t
h
e
m
ac
h
i
n
e
le
ar
n
i
n
g
a
lg
o
r
it
h
m
'
s
ap
p
li
ca
b
i
lit
y
i
n
es
ti
m
at
i
n
g
I
W
R
.
A
n
ew
ar
ea
o
f
c
o
m
p
u
t
er
s
cie
n
c
e
c
all
e
d
m
ac
h
i
n
e
le
ar
n
i
n
g
h
el
p
s
i
d
e
n
t
if
y
p
at
te
r
n
s
i
n
a
wi
d
e
r
an
g
e
o
f
f
ie
ld
s
.
Am
o
n
g
t
h
es
e,
t
h
e
f
i
el
d
s
o
f
a
g
r
ic
u
lt
u
r
e
an
d
ir
r
i
g
a
ti
o
n
ar
e
t
h
e
o
n
es
th
at
d
r
aw
r
es
ea
r
c
h
e
r
s
t
o
u
s
e
a
r
ti
f
i
cia
l
i
n
t
elli
g
e
n
ce
t
o
i
m
p
r
o
v
e
c
r
o
p
p
r
o
d
u
c
t
iv
it
y
a
n
d
o
p
t
im
i
ze
i
r
r
i
g
at
io
n
t
o
c
o
n
s
e
r
v
e
wa
te
r
.
On
e
p
o
p
u
la
r
e
n
s
e
m
b
le
-
le
a
r
n
in
g
te
ch
n
i
q
u
e
th
at
p
e
r
f
o
r
m
s
w
e
ll
o
n
v
a
r
i
o
u
s
k
i
n
d
s
o
f
d
at
a
is
g
r
ad
ie
n
t
b
o
o
s
t
in
g
r
e
g
r
ess
o
r
.
I
t
c
an
e
f
f
ec
ti
v
el
y
cl
ass
if
y
a
n
d
p
r
ed
ict
t
h
e
d
at
a.
A
d
at
ase
t
c
o
m
p
r
is
in
g
we
at
h
e
r
d
a
ta
an
d
E
T
o
v
al
u
es
is
u
s
e
d
t
o
tr
ai
n
th
e
g
r
a
d
i
en
t
b
o
o
s
tin
g
r
e
g
r
ess
o
r
al
g
o
r
it
h
m
i
n
t
h
e
s
u
g
g
es
te
d
m
o
d
el
.
T
h
e
g
r
i
d
s
e
ar
ch
m
et
h
o
d
is
u
s
e
d
to
d
ete
r
m
i
n
e
t
h
e
i
d
e
al
n
u
m
b
er
o
f
tr
ee
s
.
W
h
e
n
tes
t
d
a
ta
is
u
s
ed
,
n
e
w
E
T
c
v
a
lu
es
a
r
e
a
n
ti
ci
p
at
e
d
.
E
f
f
ec
ti
v
e
r
ai
n
is
d
e
d
u
ct
ed
f
r
o
m
th
e
p
r
e
d
i
cte
d
E
T
c
v
al
u
es
t
o
p
r
e
d
i
ct
I
W
R
v
alu
es
s
h
o
w
n
i
n
(
5
)
.
W
h
e
n
s
u
p
p
l
y
i
n
g
t
h
e
o
b
s
e
r
v
e
d
an
d
p
r
e
d
i
ct
ed
v
al
u
es
o
f
I
W
R
,
th
e
m
o
d
el'
s
a
cc
u
r
ac
y
is
v
e
r
if
ie
d
u
s
i
n
g
t
h
e
r
o
o
t
m
ea
n
s
q
u
a
r
e
d
e
r
r
o
r
(
R
MS
E
)
an
d
co
ef
f
i
cie
n
t
o
f
d
et
er
m
i
n
a
ti
o
n
(R
2
)
p
er
f
o
r
m
an
ce
m
at
r
ic
es
r
e
p
r
es
en
t
e
d
i
n
(
6
)
to
(
9
).
=
−
(
5
)
2
.
3
.
G
ra
dient
bo
o
s
t
ing
re
g
re
s
s
o
r
T
h
e
g
r
a
d
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
is
an
e
n
s
em
b
le
m
ac
h
in
e
lea
r
n
in
g
tec
h
n
iq
u
e
th
at
b
u
ild
s
d
e
cisi
o
n
tr
ee
o
n
e
af
ter
th
e
o
th
er
s
eq
u
e
n
tially
s
o
th
at
th
e
cu
r
r
en
t
o
n
e
co
r
r
ec
ts
th
e
p
r
ev
io
u
s
o
n
e'
s
er
r
o
r
.
Gr
ad
ien
t
d
ec
e
n
t
o
p
tim
izatio
n
is
its
f
o
u
n
d
atio
n
,
wh
er
e
a
u
s
er
-
d
ef
in
ed
d
if
f
e
r
en
tiab
le
lo
s
s
f
u
n
ctio
n
is
m
in
im
i
ze
d
.
Me
an
s
q
u
ar
ed
er
r
o
r
is
em
p
l
o
y
ed
as
a
l
o
s
s
f
u
n
ctio
n
f
o
r
p
r
ed
ictio
n
p
r
o
b
le
m
s
.
A
g
r
ad
ien
t
b
o
o
s
tin
g
r
e
g
r
ess
o
r
is
r
en
o
wn
e
d
f
o
r
its
h
ig
h
s
ca
lab
ilit
y
a
n
d
p
r
ec
is
io
n
.
I
t
is
ex
te
n
s
iv
ely
u
s
ed
in
a
v
ar
iety
o
f
f
ield
s
,
in
cl
u
d
in
g
h
ea
lth
r
is
k
ass
ess
m
en
t,
cr
ed
it r
is
k
ass
ess
m
en
t,
m
ed
ical
d
iag
n
o
s
is
,
r
ec
o
g
n
itio
n
o
f
p
h
o
to
s
,
an
d
tr
av
el
tim
e
esti
m
atio
n
.
T
h
is
m
o
tiv
ate
s
u
s
to
v
er
if
y
its
ap
p
licab
ilit
y
to
esti
m
ate
th
e
I
W
R
,
wh
ich
is
r
ar
ely
f
o
u
n
d
in
th
e
liter
atu
r
e.
Fo
r
th
is
,
s
ev
er
al
h
y
p
er
p
a
r
am
e
ter
s
h
av
e
b
ee
n
d
ef
i
n
ed
.
T
o
co
n
tr
o
l
th
e
lear
n
in
g
p
r
o
ce
s
s
o
f
th
e
m
o
d
el,
h
y
p
er
p
ar
am
eter
s
p
lay
an
im
p
o
r
tan
t
r
o
le.
T
h
ese
ar
e
th
e
co
n
f
ig
u
r
atio
n
v
ar
iab
les
th
at
ca
n
b
e
tu
n
ed
to
g
et
th
eir
o
p
tim
u
m
v
alu
e
to
en
h
an
ce
th
e
ef
f
icien
cy
o
f
th
e
alg
o
r
ith
m
a
n
d
av
o
id
th
e
u
n
d
er
-
f
itti
n
g
an
d
o
v
er
-
f
itti
n
g
is
s
u
es.
I
n
th
e
co
n
te
x
t
o
f
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
,
n
u
m
e
r
o
u
s
h
y
p
er
p
ar
a
m
eter
s
h
av
e
b
ee
n
d
ef
in
ed
,
s
u
ch
as
n
_
esti
m
ato
r
(
n
u
m
b
e
r
o
f
d
ec
is
i
o
n
tr
ee
s
)
,
t
h
e
d
e
p
th
o
f
th
e
tr
ee
an
d
lea
r
n
in
g
_
r
ate
,
wh
ic
h
ar
e
i
m
p
o
r
tan
t
o
n
es
th
at
in
f
lu
en
ce
th
e
ac
cu
r
ac
y
an
d
e
f
f
icien
cy
o
f
th
e
g
r
a
d
ien
t b
o
o
s
tin
g
r
eg
r
ess
o
r
alg
o
r
ith
m
.
T
h
e
n
u
m
b
e
r
o
f
tr
ee
s
(
n
_
esti
m
ato
r
)
an
d
th
e
tr
ee
'
s
d
ep
th
(
m
ax
_
d
ep
t
h
)
ar
e
two
cr
u
cial
h
y
p
er
p
ar
am
eter
s
.
I
n
th
e
cu
r
r
e
n
t
s
tu
d
y
,
we
u
s
e
th
e
g
r
id
s
ea
r
ch
s
tr
ateg
y
to
tu
n
e
t
h
e
n
_
esti
m
ato
r
wh
ile
tak
in
g
in
to
ac
co
u
n
t
th
at
th
e
m
ax
_
d
ep
th
is
5
.
T
h
e
f
o
llo
win
g
s
tep
s
s
u
m
m
ar
ize
th
e
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
m
eth
o
d
an
d
ar
e
r
ep
r
esen
ted
i
n
Fig
u
r
e
3
.
Step
1
: Bas
e
m
o
d
el
is
cr
ea
ted
b
y
p
r
e
d
ictin
g
th
e
ta
r
g
et
v
alu
e
y
with
in
itia
l
m
o
d
el
F
0
=
1
n
∑
y
i
n
i
=
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
E
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
mo
d
el
to
esti
ma
te
ir
r
ig
a
tio
n
w
a
ter r
eq
u
ir
eme
n
t
… (
S
a
te
n
d
r
a
K
u
ma
r
Ja
in
)
147
Step
2
:
Fo
r
m
=1
to
M
wea
k
lear
n
er
s
(
n
u
m
b
er
o
f
tr
ee
s
)
,
r
e
p
ea
t
f
o
llo
win
g
s
tep
s
u
n
til
th
e
d
esire
d
o
u
tco
m
e
is
r
ea
ch
ed
.
Step
3
: Res
id
u
al
is
ca
lcu
lated
b
y
r
ed
u
ci
n
g
th
e
p
r
ed
icted
v
alu
e
f
r
o
m
t
h
e
tar
g
et
v
al
u
e
r
e
s
idua
l
rs
i
=
y
i
−
F
m
−
1
Step
4
: Bu
ild
th
e
wea
k
lear
n
er
o
n
d
a
taset
D
(
x
,
rs
i
)
th
at
m
ak
es p
r
ed
icti
o
n
P
m
.
Step
5
: A
d
d
th
e
p
r
ed
ictio
n
o
f
n
ew
lear
n
er
to
th
e
p
r
ev
i
o
u
s
wea
k
lear
n
er
'
s
p
r
ed
ictio
n
to
u
p
d
ate
it.
F
m
=
F
m
−
1
+
l
e
a
r
n
i
n
g
_
r
a
te
×
P
m
Fig
u
r
e
3
.
Flo
w
ch
a
r
t o
f
g
r
ad
ie
n
t b
o
o
s
tin
g
r
eg
r
ess
o
r
m
o
d
el
2
.
4
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n
T
wo
s
tatis
tical
m
ea
s
u
r
es a
r
e
u
s
ed
in
th
is
s
tu
d
y
to
m
ea
s
u
r
e
th
e
ca
p
ab
ilit
y
o
f
p
r
o
p
o
s
ed
m
o
d
e
l:
i)
R
MSE
is
th
e
s
q
u
ar
e
r
o
o
t o
f
t
h
e
s
q
u
ar
ed
a
v
er
ag
e
o
f
r
esid
u
al
s
:
=
√
1
×
∑
(
−
̂
̂
)
2
=
1
(
6
)
ii)
R
2
is
ca
lcu
lated
as f
o
llo
ws an
d
s
h
o
ws h
o
w
d
ata
p
o
in
ts
n
ea
r
ly
lie
ac
r
o
s
s
th
e
r
eg
r
ess
io
n
lin
e
:
2
=
_
_
_
_
_
_
(
7
)
_
_
_
=
∑
(
̂
−
̅
)
2
=
1
(
8
)
_
_
_
=
∑
(
−
̅
)
2
=
1
(
9
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J
Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
142
-
1
5
4
148
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
W
h
en
ca
lcu
latin
g
a
cr
o
p
'
s
I
W
R
,
E
T
o
is
a
s
ig
n
if
ican
t
f
ac
to
r
.
E
s
tim
atin
g
I
W
R
f
o
r
th
e
wh
e
at
cr
o
p
in
L
u
d
h
i
an
a
,
Pu
n
jab
,
is
th
e
s
tu
d
y
'
s
g
o
al.
A
v
ar
iety
o
f
p
r
o
g
r
am
s
,
in
clu
d
in
g
MS
.
E
x
ce
l
a
n
d
Py
th
o
n
lib
r
ar
ies
(
Scik
it
-
lear
n
an
d
Ma
tp
lo
tlib
)
,
ar
e
em
p
lo
y
e
d
to
co
m
p
lete
th
is
task
.
Usi
n
g
th
e
FAO
-
5
6
PM
eq
u
atio
n
,
th
e
E
T
o
,
E
T
c,
an
d
I
W
R
v
alu
es
ar
e
o
b
s
er
v
ed
f
ir
s
t.
Ma
x
im
u
m
an
d
m
i
n
im
u
m
tem
p
e
r
atu
r
es,
h
u
m
id
it
y
,
win
d
s
p
ee
d
at
two
m
eter
s
ab
o
v
e
th
e
g
r
o
u
n
d
,
a
n
d
s
u
n
s
h
in
e
h
o
u
r
s
ar
e
th
e
in
p
u
t
co
lu
m
n
s
in
th
e
E
x
ce
l
s
h
e
et.
T
h
e
FAO
-
PM5
6
f
o
r
m
u
las
ar
e
u
s
ed
to
d
eter
m
in
e
th
e
v
ap
o
r
p
r
ess
u
r
e
d
ef
ici
t,
n
et
r
ad
iatio
n
,
s
h
o
r
twav
e
r
ad
iatio
n
,
lo
n
g
wa
v
e
r
ad
i
atio
n
,
an
d
ex
tr
ater
r
estrial
r
ad
iatio
n
.
T
o
esti
m
ate
th
e
E
T
o
o
f
a
ce
r
tain
r
eg
i
o
n
,
th
ese
p
ar
am
eter
s
ar
e
ess
en
tial.
T
ab
le
1
d
is
p
lay
s
th
e
o
b
s
er
v
ed
an
n
u
al
av
e
r
ag
e
E
T
o
v
alu
es.
Giv
e
n
th
at
L
u
d
h
ian
a
is
lo
ca
ted
in
a
h
o
t
,
s
em
i
-
ar
id
clim
ate.
I
n
Ma
y
an
d
J
u
n
e,
h
ig
h
E
T
o
v
alu
es a
r
e
o
b
s
er
v
ed
,
r
ea
c
h
in
g
o
v
er
1
0
m
m
p
er
d
ay
.
T
h
e
d
ataset
u
n
d
er
e
v
alu
atio
n
e
x
h
ib
its
an
n
u
al
m
ea
n
E
T
o
v
alu
es
o
f
r
o
u
g
h
ly
5
m
m
/y
ea
r
.
Daily
v
ar
iatio
n
o
f
E
T
o
is
s
h
o
wn
in
Fig
u
r
e
4
.
Fig
u
r
e
4
.
Daily
v
ar
iatio
n
in
E
T
o
v
alu
es a
t L
u
d
h
ian
a
Sin
ce
wh
ea
t
is
a
cr
o
p
th
at
g
r
o
ws
in
th
e
win
ter
.
Har
v
esti
n
g
tak
es
1
2
0
t
o
1
5
0
d
ay
s
.
T
h
er
e
ar
e
f
o
u
r
d
is
tin
ct
s
tag
es
th
r
o
u
g
h
o
u
t
th
e
wh
ea
t
cr
o
p
'
s
en
tire
g
r
o
win
g
s
ea
s
o
n
:
th
e
ea
r
ly
,
d
ev
elo
p
m
en
t,
m
id
,
an
d
late
s
tag
es.
E
ac
h
s
tag
e
h
as
a
u
n
iq
u
e
Kc
v
alu
e.
T
h
e
liter
atu
r
e
h
a
s
s
h
o
wn
th
at
th
e
Kc
v
alu
es
f
o
r
cr
o
p
s
s
o
wn
ea
r
l
y
an
d
th
o
s
e
s
o
wn
late
v
ar
y
.
T
h
e
Kc
v
alu
es
f
o
r
w
h
ea
t
cr
o
p
ar
e
tak
en
f
r
o
m
[
2
6
]
a
n
d
u
s
ed
in
th
is
in
v
esti
g
atio
n
.
T
h
e
in
h
e
r
ited
Kc
v
alu
es
ar
e
0
.
3
9
,
1
.
2
6
,
an
d
0
.
3
6
,
r
esp
ec
tiv
el
y
,
f
o
r
th
e
in
itial
s
tag
e,
m
id
-
s
ea
s
o
n
s
tag
e
,
an
d
th
e
late
-
s
ea
s
o
n
s
t
ag
e.
E
T
c
v
alu
es
f
o
r
th
e
d
ev
elo
p
m
en
t stag
e
ar
e
ca
lcu
lated
u
s
in
g
th
e
(
1
0
)
.
T
h
e
o
b
s
er
v
e
d
E
T
o
a
n
d
E
T
c
v
a
lu
es
f
o
r
th
e
d
i
f
f
er
en
t g
r
o
wth
s
t
ag
e
s
ar
e
s
h
o
wn
in
T
a
b
le
3
an
d
d
ep
icted
u
s
in
g
th
e
web
ch
ar
t
in
Fig
u
r
e
5
.
On
e
s
ig
n
if
ican
t
f
i
n
d
in
g
is
th
at
E
T
o
v
alu
es
ar
e
h
ig
h
er
th
an
E
T
c
d
u
r
in
g
th
e
f
ir
s
t,
d
ev
elo
p
m
e
n
t,
an
d
en
d
o
f
th
e
s
ea
s
o
n
.
Ho
wev
er
,
in
th
e
m
id
d
le
o
f
th
e
s
ea
s
o
n
,
th
e
o
p
p
o
s
ite
is
tr
u
e,
with
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
E
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
mo
d
el
to
esti
ma
te
ir
r
ig
a
tio
n
w
a
ter r
eq
u
ir
eme
n
t
… (
S
a
te
n
d
r
a
K
u
ma
r
Ja
in
)
149
E
T
c
v
alu
es e
x
ce
ed
in
g
E
T
o
.
I
t
in
d
icate
s
th
at
th
e
cr
o
p
lo
s
es m
o
r
e
wate
r
d
u
r
in
g
th
e
m
id
-
s
ea
s
o
n
an
d
h
as a
lo
n
g
er
life
s
p
an
,
wh
er
ea
s
E
T
o
c
o
n
tr
ib
u
tes m
o
r
e
in
th
e
in
itial,
d
ev
elo
p
m
en
t,
an
d
en
d
s
ea
s
o
n
.
_
_
ℎ
_
=
_
+
(
(
ℎ
_
−
∑
ℎ
_
)
ℎ
_
)
×
(
_
−
_
)
(
1
0
)
T
ab
le
3
.
Ob
s
er
v
e
d
E
T
o
an
d
E
T
c
v
alu
es a
t v
ar
io
u
s
wh
ea
t c
r
o
p
g
r
o
wth
s
tag
es
D
u
r
a
t
i
o
n
f
r
o
m O
c
t
28
t
o
M
a
r
21
-
22
ETo
_
i
n
i
t
i
a
l
st
a
g
e
ETc
_
i
n
i
t
i
a
l
st
a
g
e
ETo
_
d
e
v
p
st
a
g
e
ETc
_
d
e
v
p
st
a
g
e
ETo
_
mi
d
sea
s
o
n
ETc
_
mi
d
sea
s
o
n
ETo
_
e
n
d
sea
s
o
n
ETc
_
e
n
d
sea
s
o
n
2
0
0
0
-
01
4
.
2
1
.
6
3
.
3
2
.
7
3
.
6
4
.
5
5
.
6
2
.
0
2
0
0
1
-
02
4
.
2
1
.
6
3
.
0
2
.
4
3
.
3
4
.
1
5
.
1
1
.
8
2
0
0
2
-
03
3
.
7
1
.
5
3
.
1
2
.
5
3
.
0
3
.
8
4
.
8
1
.
7
2
0
0
3
-
04
3
.
7
1
.
4
2
.
8
2
.
2
3
.
0
3
.
7
6
.
1
2
.
2
2
0
0
4
-
05
3
.
6
1
.
4
3
.
0
2
.
5
2
.
6
3
.
3
4
.
3
1
.
6
2
0
0
5
-
06
4
.
0
1
.
6
3
.
0
2
.
5
3
.
6
4
.
5
4
.
9
1
.
8
2
0
0
6
-
07
3
.
4
1
.
3
2
.
5
2
.
0
2
.
9
3
.
6
4
.
0
1
.
5
2
0
0
7
-
08
3
.
9
1
.
5
3
.
0
2
.
5
3
.
1
3
.
9
5
.
8
2
.
1
2
0
0
8
-
09
3
.
4
1
.
3
2
.
7
2
.
2
2
.
9
3
.
7
5
.
4
1
.
9
2
0
0
9
-
10
3
.
6
1
.
4
2
.
8
2
.
3
3
.
2
4
.
1
5
.
5
2
.
0
2
0
1
0
-
11
3
.
4
1
.
3
2
.
3
1
.
9
2
.
8
3
.
6
4
.
8
1
.
7
2
0
1
1
-
12
3
.
3
1
.
3
2
.
5
2
.
0
2
.
8
3
.
5
5
.
3
1
.
9
2
0
1
2
-
13
3
.
5
1
.
4
2
.
9
2
.
3
2
.
9
3
.
6
4
.
7
1
.
7
2
0
1
3
-
14
3
.
3
1
.
3
2
.
8
2
.
3
2
.
8
3
.
5
4
.
5
1
.
6
2
0
1
4
-
15
3
.
6
1
.
4
2
.
7
2
.
1
2
.
8
3
.
6
3
.
8
1
.
4
2
0
1
5
-
16
3
.
5
1
.
4
2
.
8
2
.
4
3
.
1
3
.
9
5
.
0
1
.
8
2
0
1
6
-
17
3
.
6
1
.
4
3
.
2
2
.
7
2
.
8
3
.
5
4
.
8
1
.
7
2
0
1
7
-
18
3
.
1
1
.
2
2
.
8
2
.
3
3
.
2
4
.
0
5
.
1
1
.
8
2
0
1
8
-
19
3
.
0
1
.
2
2
.
3
1
.
8
2
.
3
2
.
9
3
.
7
1
.
3
2
0
1
9
-
20
3
.
0
1
.
2
1
.
9
1
.
6
2
.
3
2
.
9
3
.
7
1
.
3
2
0
2
0
-
21
3
.
4
1
.
3
2
.
4
2
.
0
2
.
7
3
.
4
5
.
6
2
.
0
Fig
u
r
e
5
.
R
elatio
n
s
h
ip
b
etwe
e
n
E
T
o
a
n
d
E
T
c
i
n
th
e
in
itial,
d
ev
elo
p
m
en
t,
m
id
,
a
n
d
late
s
ea
s
o
n
C
r
o
p
g
r
o
wth
an
d
p
r
o
d
u
ctiv
it
y
ar
e
im
p
ac
ted
b
y
cr
o
p
wate
r
s
tr
ess
,
wh
ich
is
wh
en
a
p
lan
t
d
o
es
n
o
t
r
ec
eiv
e
en
o
u
g
h
wate
r
to
g
r
o
w.
I
r
r
ig
atio
n
an
d
r
ain
f
all
ca
n
h
elp
r
ed
u
ce
it.
So
m
e
p
r
ec
ip
i
tatio
n
-
r
elate
d
wate
r
ev
ap
o
r
ates
a
n
d
r
u
n
s
o
f
f
.
T
h
e
s
o
il
d
o
es n
o
t
a
b
s
o
r
b
all
o
f
th
e
p
r
ec
ip
ita
tio
n
,
an
d
th
e
cr
o
p
d
o
es
n
o
t u
s
e
it
to
g
r
o
w.
E
f
f
ec
tiv
e
wate
r
is
th
e
wate
r
th
at
is
ab
s
o
r
b
ed
b
y
th
e
s
o
il a
n
d
u
tili
ze
d
b
y
th
e
c
r
o
p
f
o
r
g
r
o
wth
.
I
n
th
is
r
esear
ch
,
we
ex
a
m
in
e
2
1
wh
ea
t
-
cr
o
p
d
u
r
atio
n
s
(
p
lan
tin
g
to
h
a
r
v
esti
n
g
)
f
r
o
m
2
0
0
0
to
2
0
2
1
.
T
ab
le
4
lis
ts
to
tal
r
ain
f
all,
ef
f
ec
tiv
e
r
ain
f
all,
o
b
s
er
v
e
d
E
T
c,
an
d
o
b
s
er
v
e
d
I
W
R
in
f
o
r
m
atio
n
y
ea
r
ly
f
o
r
ea
ch
wh
ea
t
-
cr
o
p
-
d
u
r
atio
n
.
I
t
ca
n
b
e
n
o
ticed
i
n
th
e
ta
b
le
th
at
th
e
g
r
ea
test
am
o
u
n
t
o
f
p
r
ec
ip
itatio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
8
1
4
I
n
t J
Ad
v
Ap
p
l Sci
,
Vo
l.
1
5
,
No
.
1
,
Ma
r
c
h
2
0
2
6
:
142
-
1
5
4
150
(
2
0
7
.
3
m
m
/
s
ea
s
o
n
)
o
cc
u
r
r
ed
b
etwe
en
Octo
b
er
2
8
,
2
0
1
9
,
an
d
Ma
r
ch
2
1
,
2
0
2
0
.
T
h
e
E
T
c
v
alu
e
o
b
s
er
v
ed
f
o
r
th
is
p
er
io
d
is
2
8
0
.
2
m
m
/s
ea
s
o
n
,
wh
er
ea
s
ef
f
ec
tiv
e
r
ain
is
co
m
p
u
ted
as
1
8
0
.
7
m
m
/s
ea
s
o
n
.
T
h
e
lo
west
am
o
u
n
t
o
f
p
r
ec
ip
itatio
n
(
2
1
.
3
m
m
/
s
ea
s
o
n
)
o
cc
u
r
r
ed
b
etwe
en
Octo
b
er
2
8
,
2
0
0
0
,
an
d
Ma
r
c
h
2
2
,
2
0
0
1
.
Ob
s
er
v
e
d
E
T
c
v
alu
e
f
o
r
th
is
p
er
io
d
is
4
3
9
.
2
m
m
/s
ea
s
o
n
,
wh
er
ea
s
th
e
co
m
p
u
ted
ef
f
ec
tiv
e
r
ain
is
2
0
.
6
.
T
h
e
I
W
R
v
alu
es
f
o
r
th
ese
p
er
io
d
s
ar
e
9
9
.
6
a
n
d
4
1
8
.
6
,
r
esp
ec
tiv
ely
.
An
in
te
r
e
s
tin
g
f
in
d
in
g
is
th
at
cr
o
p
ir
r
ig
atio
n
q
u
a
n
tity
is
in
f
lu
en
ce
d
b
y
ef
f
ec
tiv
e
r
ain
f
a
ll.
L
o
w
e
f
f
ec
tiv
e
r
ain
f
all,
wh
ich
r
esu
lts
f
r
o
m
less
r
ain
f
all,
n
ec
ess
itates
m
o
r
e
wate
r
f
o
r
ir
r
i
g
atio
n
,
wh
e
r
ea
s
h
ig
h
ef
f
ec
tiv
e
r
ain
f
all,
wh
ich
r
e
s
u
lts
f
r
o
m
m
o
r
e
r
ain
f
all,
r
e
q
u
i
r
es less
wate
r
.
T
ab
le
4
.
Seaso
n
al
I
W
R
f
o
r
wh
ea
t in
L
u
d
h
ia
n
a
D
u
r
a
t
i
o
n
f
r
o
m
O
c
t
28
t
o
M
a
r
21
-
22
To
t
a
l
r
a
i
n
(
mm
)
Ef
f
e
c
t
i
v
e
r
a
i
n
(
mm
)
O
b
serv
e
d
ET
c
(
mm
/
sea
s
o
n
)
O
b
serv
e
d
I
W
R
(
mm
/
sea
s
o
n
)
2
0
0
0
-
01
2
1
.
3
2
0
.
6
4
3
9
.
2
4
1
8
.
6
2
0
0
1
-
02
4
3
.
4
4
1
.
1
4
0
2
.
1
3
6
1
.
0
2
0
0
2
-
03
1
1
4
.
0
9
3
.
4
3
8
0
.
4
2
8
7
.
0
2
0
0
3
-
04
6
4
.
5
5
7
.
4
3
7
9
.
9
3
2
2
.
5
2
0
0
4
-
05
1
6
0
.
4
1
4
1
.
0
3
4
8
.
9
2
2
7
.
9
2
0
0
5
-
06
5
2
.
0
4
8
.
1
4
2
1
.
8
3
7
9
.
7
2
0
0
6
-
07
1
7
3
.
8
1
4
1
.
3
3
4
0
.
9
2
2
8
.
7
2
0
0
7
-
08
3
4
.
2
3
2
.
8
4
0
2
.
2
3
6
9
.
3
2
0
0
8
-
09
6
1
.
4
5
7
.
9
3
6
9
.
4
3
1
1
.
5
2
0
0
9
-
10
2
5
.
4
2
4
.
8
3
9
5
.
4
3
7
0
.
6
2
0
1
0
-
11
5
4
.
2
5
0
.
1
3
4
3
.
1
2
9
3
.
0
2
0
1
1
-
12
3
7
.
5
3
5
.
7
3
4
8
.
2
3
1
2
.
5
2
0
1
2
-
13
1
2
5
.
2
1
1
4
.
1
3
6
2
.
0
2
6
3
.
8
2
0
1
3
-
14
8
5
.
2
7
9
.
8
3
5
3
.
2
2
7
3
.
4
2
0
1
4
-
15
1
9
1
.
9
1
5
3
.
8
3
4
4
.
7
2
3
3
.
4
2
0
1
5
-
16
6
9
.
4
6
2
.
5
3
8
2
.
2
3
3
1
.
4
2
0
1
6
-
17
1
1
1
.
2
9
7
.
4
3
7
4
.
7
2
8
1
.
6
2
0
1
7
-
18
7
3
.
5
6
7
.
8
3
8
2
.
0
3
1
5
.
3
2
0
1
8
-
19
1
5
3
.
0
1
3
1
.
3
2
9
4
.
7
1
6
6
.
2
2
0
1
9
-
20
2
0
7
.
3
1
8
0
.
7
2
8
0
.
2
1
5
1
2
0
2
0
-
21
6
3
.
1
5
9
.
9
3
4
6
.
3
2
8
7
.
9
A
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
,
co
u
ld
b
e
a
r
e
p
lace
m
en
t
o
f
FAO
-
5
6
PM
ap
p
r
o
ac
h
to
p
r
e
d
ict
th
e
I
W
R
.
As
s
h
o
wn
in
T
ab
le
5
,
th
e
d
ata
s
et,
wh
ich
co
m
p
r
is
es
8
0
3
6
s
a
m
p
les
(
s
p
an
n
in
g
th
e
y
ea
r
s
2
0
0
0
–
2
0
2
1
)
,
is
d
iv
id
ed
in
to
tr
ain
in
g
(
7
8
9
0
s
am
p
les)
an
d
test
(
1
4
6
s
am
p
les)
d
ataset.
T
h
e
test
d
ataset
in
clu
d
es
s
am
p
les
o
f
wh
ea
t
cr
o
p
s
g
r
o
wn
f
r
o
m
Octo
b
er
2
8
,
2
0
2
0
,
to
Ma
r
ch
2
2
,
2
0
2
1
.
T
h
e
g
r
id
s
ea
r
ch
ap
p
r
o
ac
h
is
u
s
ed
in
th
is
s
tu
d
y
to
in
itia
lly
d
eter
m
in
e
th
e
b
est
v
alu
e
o
f
th
e
h
y
p
er
p
ar
a
m
eter
,
n
_
esti
m
ato
r
(
n
u
m
b
e
r
o
f
d
ec
is
io
n
tr
ee
)
.
T
h
e
m
ea
n
test
s
co
r
e
o
f
h
y
p
e
r
p
ar
am
eter
tu
n
i
n
g
is
r
ep
r
esen
ted
in
T
ab
le
6
.
T
h
e
b
est
av
ailab
le
h
y
p
er
p
ar
am
eter
is
u
s
ed
to
tr
ain
th
e
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
u
s
in
g
a
tr
ain
in
g
d
a
taset.
T
h
e
m
o
d
el’
s
p
r
ed
ictio
n
is
s
u
m
m
ar
ize
d
in
T
ab
le
7
f
o
r
th
e
test
d
at
a
s
et.
Fo
r
R
MSE
an
d
R
2
,
th
e
g
r
ad
ien
t
b
o
o
s
tin
g
r
eg
r
ess
o
r
p
r
e
d
icts
E
T
o
v
alu
es
with
an
a
cc
u
r
ac
y
o
f
0
.
2
2
m
m
/d
ay
an
d
0
.
9
8
,
r
esp
ec
tiv
ely
.
R
eg
r
ess
io
n
an
aly
s
is
b
et
wee
n
o
b
s
er
v
ed
E
T
o
a
n
d
p
r
e
d
i
cted
E
T
o
is
s
h
o
wn
in
Fig
u
r
e
6
.
Fig
u
r
e
6
(
a)
p
r
ese
n
ts
th
e
r
e
g
r
ess
io
n
lin
e
Y
=1
×
X+
0
.
0
8
,
wh
ile
th
e
co
r
r
esp
o
n
d
in
g
r
esid
u
al
g
r
ap
h
is
s
h
o
wn
in
Fig
u
r
e
6
(
b
)
.
T
h
e
Kc
an
d
p
r
ed
icted
E
T
o
v
alu
es
ar
e
tak
en
in
to
ac
co
u
n
t
wh
en
p
r
e
d
ictin
g
E
T
c
v
alu
es.
T
h
e
g
r
ad
ien
t
b
o
o
s
tin
g
r
e
g
r
ess
o
r
p
r
ed
icts
E
T
c
v
alu
es
with
an
ac
cu
r
ac
y
o
f
0
.
1
9
m
m
/d
ay
a
n
d
0
.
9
8
f
o
r
R
MSE
an
d
R
2
,
r
esp
ec
tiv
ely
.
R
eg
r
ess
io
n
an
aly
s
is
b
etwe
en
o
b
s
er
v
e
d
E
T
c
an
d
p
r
ed
icted
E
T
c
is
s
h
o
wn
in
Fig
u
r
e
7
.
Fig
u
r
e
7
(
a)
p
r
esen
ts
th
e
r
e
g
r
ess
io
n
lin
e
Y=
0
.
9
7
×X
+0
.
1
2
,
wh
ile
th
e
c
o
r
r
esp
o
n
d
in
g
r
esid
u
al
g
r
a
p
h
is
r
ep
r
esen
ted
in
Fig
u
r
e
7
(
b
)
.
Fi
n
ally
,
I
W
R
is
esti
m
ated
wi
th
an
ac
cu
r
ac
y
o
f
0
.
1
8
m
m
/d
ay
an
d
0
.
9
8
f
o
r
R
MSE
an
d
R
2
,
r
esp
ec
tiv
ely
.
R
eg
r
ess
io
n
an
aly
s
is
b
etwe
en
o
b
s
er
v
ed
I
W
R
an
d
p
r
ed
icted
I
W
R
is
s
h
o
wn
in
Fig
u
r
e
8
.
Fig
u
r
e
8
(
a)
p
r
esen
ts
th
e
r
e
g
r
ess
io
n
lin
e
Y=
0
.
9
8
×X
+0
.
1
1
,
wh
ile
th
e
c
o
r
r
esp
o
n
d
in
g
r
esid
u
al
g
r
a
p
h
is
r
ep
r
esen
ted
in
Fig
u
r
e
8
(
b
)
.
T
a
b
le
7
s
h
o
ws
th
e
r
elatio
n
s
h
ip
b
etwe
en
o
b
s
er
v
ed
I
W
R
an
d
p
r
e
d
icted
I
W
R
.
I
t
ca
n
b
e
n
o
ticed
th
at
f
o
r
th
e
wh
ea
t
cr
o
p
s
g
r
o
wn
f
r
o
m
Octo
b
er
2
8
,
2
0
2
0
,
t
o
Ma
r
c
h
2
2
,
2
0
2
1
,
th
e
o
b
s
er
v
e
d
I
W
R
is
2
8
7
.
9
m
m
/s
ea
s
o
n
,
w
h
er
ea
s
p
r
ed
icted
I
W
R
is
2
9
6
.
0
m
m
/s
ea
s
o
n
.
T
h
e
a
cc
u
r
ac
y
d
em
o
n
s
tr
ated
b
y
g
r
a
d
ien
t
b
o
o
s
tin
g
r
e
g
r
ess
o
r
is
s
u
m
m
ar
i
ze
d
in
T
ab
le
8
.
T
ab
le
5
.
A
d
i
v
is
io
n
o
f
t
r
ain
in
g
an
d
test
d
ata
D
a
t
a
s
e
t
d
u
r
a
t
i
o
n
To
t
a
l
sam
p
l
e
s
Tr
a
i
n
i
n
g
d
a
t
a
se
t
Te
st
d
a
t
a
set
01
-
J
a
n
-
2
0
0
0
t
o
3
1
-
D
e
c
-
2
0
2
1
8
,
0
3
6
7
,
8
9
0
(
0
1
-
Ja
n
-
2
0
0
t
o
2
7
-
O
c
t
2
0
2
0
)
a
n
d
(
2
3
-
M
a
r
-
2
0
2
1
t
o
3
1
-
D
e
c
-
2
0
2
1
)
1
4
6
(
2
8
-
O
c
t
-
2
0
2
0
t
o
2
2
-
M
a
r
-
2
0
2
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2252
-
8
8
1
4
E
n
s
emb
le
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
mo
d
el
to
esti
ma
te
ir
r
ig
a
tio
n
w
a
ter r
eq
u
ir
eme
n
t
… (
S
a
te
n
d
r
a
K
u
ma
r
Ja
in
)
151
T
ab
le
6
.
R
esu
lts
o
f
ad
ju
s
tin
g
h
y
p
er
p
a
r
am
eter
s
f
o
r
v
ar
y
i
n
g
tr
e
e
co
u
n
ts
n
_
e
st
i
ma
t
o
r
10
20
30
40
50
60
70
80
90
1
0
0
max
-
d
e
p
t
h
5
5
5
5
5
5
5
5
5
5
M
e
a
n
t
e
s
t
s
c
o
r
e
0
.
8
3
4
0
.
9
4
9
0
.
9
6
8
0
.
9
7
3
0
.
9
7
5
0
.
9
7
6
0
.
9
7
7
0
.
9
7
7
0
.
9
7
7
0
.
9
7
7
T
ab
le
7
.
Pre
d
icted
I
W
R
f
o
r
w
h
ea
t c
r
o
p
i
n
L
u
d
h
ian
a
2
8
O
c
t
2
0
2
0
t
o
2
2
M
a
r
2
0
2
1
D
e
c
a
d
e
R
a
i
n
(
mm
/
d
e
c
.
)
Ef
f
e
c
t
i
v
e
r
a
i
n
(
mm
/
d
e
c
.
)
O
b
serv
e
d
ET
c
(
mm
/
d
e
c
.
)
P
r
e
d
i
c
t
e
d
ET
c
(
mm
/
d
e
c
.
)
O
b
serv
e
d
I
W
R
(
mm
/
d
e
c
.
)
P
r
e
d
i
c
t
e
d
I
W
R
(
mm
/
d
e
c
.
)
O
c
t
1
0
.
0
0
.
0
6
.
4
7
.
1
6
.
4
7
.
1
N
o
v
1
0
.
0
0
.
0
1
3
.
9
1
4
.
1
1
3
.
8
1
4
.
0
2
1
3
.
6
1
2
.
7
1
1
.
3
1
1
.
4
0
.
0
0
.
0
3
6
.
6
6
.
3
1
3
.
1
1
4
.
0
6
.
8
7
.
6
D
e
c
1
0
.
5
0
.
5
1
9
.
1
2
0
.
6
1
8
.
6
2
0
.
1
2
8
.
4
8
.
1
2
0
.
6
2
1
.
3
1
2
.
5
1
3
.
3
3
3
.
9
3
.
8
2
6
.
7
2
7
.
8
2
2
.
9
2
4
.
0
Jan
1
1
7
.
5
1
6
.
1
2
1
.
4
2
4
.
4
5
.
4
8
.
3
2
0
.
0
0
.
0
2
9
.
2
2
9
.
5
2
9
.
2
2
9
.
5
3
1
.
1
1
.
1
3
7
.
3
3
3
.
9
3
6
.
2
3
2
.
8
F
e
b
1
5
.
6
5
.
5
3
9
.
0
4
1
.
5
3
3
.
5
3
6
.
1
2
0
.
0
0
.
0
4
4
.
8
4
5
.
2
4
4
.
8
4
5
.
2
3
1
.
1
1
.
1
1
7
.
2
1
7
.
5
1
6
.
1
1
6
.
5
M
a
r
1
1
.
2
1
.
2
2
0
.
2
2
0
.
6
1
9
.
0
1
9
.
4
2
0
.
7
0
.
7
2
1
.
4
2
1
.
3
2
0
.
7
2
0
.
6
3
2
.
8
2
.
8
4
.
9
5
.
1
2
.
1
2
.
3
T
ab
le
8
.
Gr
a
d
ien
t b
o
o
s
tin
g
r
eg
r
ess
io
n
's ac
cu
r
ac
y
in
p
r
ed
ictin
g
E
T
o
,
E
T
c,
an
d
I
W
R
P
a
r
a
me
t
e
r
ETo
ETc
I
W
R
R
M
S
E
0
.
2
2
mm
/
d
a
y
0
.
1
9
mm
/
d
a
y
0
.
1
8
mm
/
d
a
y
R
2
0
.
9
8
0
.
9
8
0
.
9
8
(
a)
(
b
)
Fig
u
r
e
6
.
R
elatio
n
s
h
ip
b
etwe
e
n
o
b
s
er
v
e
d
an
d
p
r
e
d
icted
E
T
o
:
(
a)
r
eg
r
ess
io
n
lin
e
Y=
1
×X
+0
.
0
8
an
d
(
b
)
co
r
r
esp
o
n
d
in
g
r
esid
u
al
g
r
a
p
h
(
a)
(
b
)
Fig
u
r
e
7
.
R
elatio
n
s
h
ip
b
etwe
e
n
o
b
s
er
v
e
d
an
d
p
r
e
d
icted
E
T
c
:
(
a)
r
eg
r
ess
io
n
lin
e
Y=
0
.
9
7
×X
+0
.
1
2
a
n
d
(
b
)
co
r
r
esp
o
n
d
in
g
r
esid
u
al
g
r
a
p
h
Evaluation Warning : The document was created with Spire.PDF for Python.