Cul i a ing
esilience
by
empi ically
e ealing
esponse
di e si y
Helena
Kahiluo o
a,
*,
Janne
Kase a
b
,
Kaija
Hakala
b
,
Sa i
J.
Himanen
a
,
Lau i
Jauhiainen
b
,
Reimund
P.
Ro
¨ e
a
,
Tapio
Salo
b
,
Mi ek
T nka
c,d
a
MTT
Ag i ood
Resea ch
Finland,
Plan
P oduc ion
Resea ch,
Lo
¨nn o inka u
5,
FI-50100
Mikkeli,
Finland
b
MTT
Ag i ood
Resea ch
Finland,
Plan
P oduc ion
Resea ch,
FI-31600
Jokioinen,
Finland
c
Ins i u e
o
Ag osys ems
and
Bioclima ology,
Mendel
Uni e si y
in
B no,
Zemedelska
1,
613
00
B no,
Czech
Republic
d
CzechGlobe
–
Global
Change
Resea ch
Cen e
AV
CR,
. .i.,
Be
ˇlidla
986/4a,
603
00
B no,
Czech
Republic
1.
In oduc ion
In ensified
clima e
and
ma ke
u bulence
has
b ough
conside able
unce ain y
o
human
ac i i ies
(Coumou
and
Rahms o ,
2012;
Dessai
e
al.,
2007).
The
ola ili y
o
he
ood
and
financial
ma ke s
has
ein oduced
ood
secu i y
on
o
he
wo ld
agenda.
Resilience
and
adap i e
capaci y,
obus ness
and
mul i-s abili y
a e
equi ed
o
complemen
he
‘p edic
and
adap ’
app oach
o
p epa ing
o
p ojec ed
long- e m
changes
(Dessai
e
al.,
2007;
Sche e
e
al.,
2001).
Di e sifica ion
is
he
s a egy
wi h
highes
expec a ions,
wi h
esponse
di e si y
being
he
key
(Folke
e
al.,
2004;
Elmq is
e
al.,
2003).
Response
di e si y,
i
empi ically
assessed,
could
lay
he
g oundwo k
o
adap i e
managemen
and
acili a e,
a
he
in e aces
o
science,
policy
and
p i a e
ac o s,
adap i e
go e nance
o
a
esilien
socie y.
To
ecognise
esilience,
we
mus
mo e
beyond
species,
cul i a
and
gene ic
di e si y.
Di e si y
in
unc ional
p ope ies
a he
han
di e si y
o
ypes
pe
se
(Page,
2010)
is
c ucial
o
he
p o ision
o
ecosys em
se ices
(Diaz
e
al.,
2007).
Response
di e si y
e e s
o
he
di e si y
o
esponses
wi hin
a
unc ional
g oup
(e.g.
wi hin
a
species,
o
g oup
o
species
p o iding
he
same
unc ion)
(Elmq is
e
al.,
2003;
Nys o
¨m,
2006).
While
p o iding
di e si y
o
esponses
o
dis u bances,
esponse
di e si y
wi hin
a
unc ional
g oup
ensu es
ha
a
leas
some
membe s
o
he
g oup
main ain
hei
unc ion
when
acing
such
dis u bances.
Consequen ly,
esponse
di e si y
enables
he
con inuous
p o ision
o
he
same
unc ion
in
u bulen
and
changing
en i onmen s
also
(Folke
e
al.,
2004;
Nys o
¨m,
2006).
In
addi ion,
esponse
di e si y,
by
p o iding
ma e ial
o
selec ion
in
new
condi ions
o
o
new
a ge s,
builds
Global
En i onmen al
Change
25
(2014)
186–193
A
R
T
I
C
L
E
I
N
F
O
A icle
his o y:
Recei ed
8
Feb ua y
2013
Recei ed
in
e ised
o m
18
Janua y
2014
Accep ed
3
Feb ua y
2014
A ailable
online
4
Ma ch
2014
Keywo ds:
Gene ic
app oach
Clima e
change
Food
secu i y
Ag i ood
sys ems
Cul i a s
Adap i e
capaci y
A
B
S
T
R
A
C
T
In ensified
clima e
and
ma ke
u bulence
equi es
esilience
o
a
mul i ude
o
changes.
Di e si y
educes
he
sensi i i y
o
dis u bance
and
os e s
he
capaci y
o
adap
o
a ious
u u e
scena ios.
Wha
eally
ma e s
is
di e si y
o
esponses.
Despi e
appeals
o
manage
esilience,
concep ual
de elopmen s
ha e
no
ye
yielded
a
b eak- h ough
in
empi ical
applica ions.
He e,
we
p esen
an
app oach
o
empi ically
e eal
he
‘ esponse
di e si y’:
he
ac o s
o
change
ha
a e
c i ical
o
a
sys em
a e
iden ified,
and
he
esponse
di e si y
is
de e mined
based
on
he
documen ed
componen
esponses
o
hese
ac o s.
We
illus a e
his
app oach
and
i s
added
alue
using
an
example
o
secu ing
ood
supply
in
he
ace
o
clima e
a iabili y
and
change.
This
example
demons a es
ha
quan i ying
esponse
di e si y
allows
o
a
new
pe spec i e:
despi e
con inued
inc ease
in
cul i a
di e si y
o
ba ley,
he
di e si y
in
esponses
o
wea he
declined
du ing
he
las
decade
in
he
egions
whe e
mos
o
he
ba ley
is
g own
in
Finland.
This
was
due
o
g ea e
homogenei y
in
esponses
among
new
cul i a s
han
among
olde
ones.
Such
a
decline
in
he
esponse
di e si y
indica es
inc eased
ulne abili y
and
educed
esilience.
The
assessmen
se es
adap i e
managemen
in
he
ace
o
bo h
ecological
and
socio-
economic
d i e s.
Supplie
di e si y
in
he
ood
e ail
indus y
in
o de
o
secu e
a o dable
ood
in
spi e
o
global
p ice
ola ili y
could
ep esen
ano he
applica ion.
The
app oach
is,
indeed,
applicable
o
any
sys em
o
which
i
is
possible
o
adop
empi ical
in o ma ion
ega ding
he
esponse
by
i s
componen s
o
he
c i ical
ac o s
o
a iabili y
and
change.
Ta ge ing
di e sifica ion
in
esponse
o
c i ical
change
b ings
e ficiency
in o
di e si y.
We
p opose
he
gene ic
p ocedu e
ha
is
demons a ed
in
his
s udy
as
a
means
o
e ficien ly
enhance
esilience
a
mul iple
le els
o
ag i ood
sys ems
and
beyond.
ß
2014
The
Au ho s.
Published
by
Else ie
L d.
*Co esponding
au ho .
Tel.:
+358
405118335;
ax:
+358
20772040.
E-mail
add esses:
helena.kahiluo o@m .fi
(H.
Kahiluo o),
Janne.Kase a@m .fi
(J.
Kase a),
Kaija.Hakala@m .fi
(K.
Hakala),
Sa i.Himanen@m .fi
(S.J.
Himanen),
Lau i.Jauhiainen@m .fi
(L.
Jauhiainen),
Reimund.Ro e @m .fi
(R.P.
Ro
¨ e ),
Tapio.Salo@m .fi
(T.
Salo),
[email p o ec ed]
(M.
T nka).
Con en s
lis s
a ailable
a
ScienceDi ec
Global
En i onmen al
Change
jo
u
n
al
h
o
mep
ag
e:
www
.else ie
.co
m
/loc
a e/g
lo
en c
h
a
h p://dx.doi.o g/10.1016/j.gloen cha.2014.02.002
0959-3780 ß
2014
The
Au ho s.
Published
by
Else ie
L d.
Open access unde CC BY-NC-ND license.
Open access unde CC BY-NC-ND license.
he
capaci y
o
success ul
ans o ma ions
(Chapin
e
al.,
1997).
The e o e,
heo e ically,
di e si y
does
no
pe
se
enhance
esilience,
whe eas
di e si y
in
esponses
o
c i ical
a iabili y
and
change
p oduces
such
enhancemen .
Despi e
appeals
o
manage
o
esilience
(Folke
e
al.,
2004;
Chapin
e
al.,
1997;
Sche e
e
al.,
2001),
he
concep ual
and
heo e ical
de elopmen
o
his
app oach
has
gene a ed
ew
empi ical
applica ions
o
da e
(Lalibe e
e
al.,
2010).
A
limi ed
numbe
o
field
s udies
ha e
obse ed
ha
esponse
di e si y
se es
o
sus ain
sys em
unc ions
ollowing
dis u bances
in
co al
ee s
(Nys o
¨m,
2006),
lakes
(Schindle ,
1990),
bee
communi ies
(Win ee
and
K emen,
2009),
ice
fields
(Zhu
e
al.,
2000)
and
g asslands
(Walke
e
al.,
1999).
Indi ec
assessmen s
o
he
impac
o
managemen
on
esponse
di e si y,
which
depend
on
he
gene ic
and
hypo he ical
di ision
o
plan
unc ion
and
esponse
ai s,
ha e
also
been
epo ed
(Lalibe e
e
al.,
2010).
Howe e ,
he
adequa e
classifica ion
o
esponses
should
be
based
on
he
unc ion
o
in e es
(Aubin
e
al.,
2009)
and
eflec
di e en ial
esponses
o
oughly
specified
c i ical
dis u bances
(Naeem
and
W igh ,
2003).
In
an
ag i ood
sys em,
he
esponse
ai s
o
odde
and
ood
supply
may
be
di e en
o
shi s
in,
o
example,
clima e
and
pes s,
demand
and
p ice,
e en
a
he
cul i a
le el.
The e o e,
he
esponse
di e si y
mus
be
iden ified
and
quan ified
di ec ly
(Aubin
e
al.,
2009)
o
each
gi en
ques ion
and
case
(Pe chey
and
Gas on,
2006).
Mul i a ia e
s a is ical
me hods,
including
clus e -
ing
and
o dina ion
me hods
ha
a e
applied
o
assess
gene ic
o
species
di e si y
(Lalibe e
e
al.,
2010;
Pe chey
and
Gas on,
2006;
Mohammadi
and
P asanna,
2003),
p o ide
examples
o
me hodo-
logical
solu ions
o
he
di ec
empi ical
quan ifica ion
o
esponse
di e si y.
He e,
we
in oduce
an
empi ical
app oach
o
di ec ly
e ealing
esponse
di e si y
and
apply
his
app oach
o
a
case
o
ood
secu i y
when
acing
clima e
change,
i.e.
o
ba ley
cul i a
esponses
o
wea he
in
Finland.
Ba ley
cul i a s
a y
in
esponse
o
wea he
pa ame e s
(Hakala
e
al.,
2012).
Fo
example,
pa icula
cul i a s
a e
d ough
suscep ible,
whe eas
o he s
do
no
ole a e
flooding
o
hea
s ess.
We
hypo hesised
ha
he
assessmen
o
he
esponse
di e si y
would
yield
a
di e en
es ima e
o
he
egional
cul i a
di e si y
han
ha
ob ained
om
me e
ype
di e si y.
I
so,
hen
he
app oach
based
on
esponse
di e si y
would
allow
a
mo e
alid
assessmen
o
di e si y
in
e ms
o
he
esponse
o
clima e
a iabili y
and
change.
In
he
case
o
added
alue
by
esponse
di e si y,
his
app oach
could
p o ide
a
gene ic
p ocedu e
as
a
p ac ical
ool
o
manage
esilience.
2.
Ma e ials
and
me hods
Ou
analysis
in ol ed
wo
s ages
ha
we e
composed
o
fi e
s eps
(Fig.
1).
2.1.
S age
I:
Iden ifica ion
o
he
esponses
o
change
ac o s
S age
I
de e mines
he
ac o s
o
change
ha
a e
c i ical
o
he
sys em
pe o mance
and
he
componen
esponses
o
a ia ions
in
hese
ac o s.
In
ou
example,
we
conside ed
he
ag o-clima ic
pa ame e s
mos
c i ical
o
ba ley
g ain
yield
(Hakala
e
al.,
2012;
Ro
¨ e
e
al.,
2013;
T nka
e
al.,
2011)
and
he
g ain
yield
esponse
o
ba ley
cul i a s
o
a ia ions
in
hese
pa ame e s
in
mul i-
loca ion
ials
(Hakala
e
al.,
2012),
which
spanned
h ee
decades,
in
Finland.
The
gene ali y
o
he
esul s
can
be
es ed
by
alida ing
he
c i ical
change
ac o s
and
esponses
using
o he
da a.
We
de e mined
he
co ela ion
in
cul i a
esponses
be ween
he
ial
da a
and
da a
om
a ms
o
es ,
whe he
he
cul i a s
espond
o
he
ag o-clima ic
pa ame e s
unde
a m
condi ions
simila ly
as
in
he
ials,
i.e.
whe he
he
esponse
di e si y
model
ha
was
c ea ed
using
he
ial
da a
is
alid
in
p ac ical
a ming
condi ions,
and
hus
applicable
o
guide
he
adap i e
managemen
o
a me s
and
decision-making
in,
o
example,
b eeding
o
ag icul u al
policy.
2.1.1.
S ep
1:
selec ing
he
c i ical
ac o s
o
change
and
a ia ion
Da a
om
he
MTT
Ag i ood
Resea ch
Finland
O ficial
Va ie y
T ials
(Hakala
e
al.,
2012)
om
14
loca ions
om
Mie oinen
in
he
sou h
(60823
0
N,
22833
0
E)
o
Ruukki
in
he
no h
(64840
0
N,
25806
0
E)
and
o
Tohmaja
¨ i
in
he
eas
(62814
0
N,
30821
0
E)
we e
used.
Consequen ly,
he
cul i a
ials
ep esen ed
all
o
Finland
excep
o
he
no he nmos
pa
o
Lapland,
i.e.
o
egion
I,
and
he
sou h-
wes e n
peninsula
o
Ah enanmaa,
i.e.
egion
XVI
(Table
2,
Fig.
2).
Six
ials
we e
in
egions
II
o
VIII
and
eigh
ials
we e
in
egions
IX
o
XV
(Fig.
2).
The
ials
we e
o
a
andomised
comple e
block
design
o
an
incomple e
block
design.
The
numbe
o
eplica es
was
3
o
4.
Cul i a s
in
he
expe imen s
di e ed
in
he
long
e m;
howe e ,
s anda d
e e ence
cul i a s
we e
used
ac oss
he
ials.
Fe ilize
use
depended
on
he
c opping
his o y,
soil
ype
and
soil
e ili y
and
was
consis en
wi h
he
a me
p ac ices
(Hakala
e
al.,
2012).
Cul i a s
o
which
he e
we e
mo e
han
25
obse a ions
we e
included
in
he
analysis.
Es ima es
we e
subs i u ed
o
a
ew
missing
alues
o
he
phenological
de elopmen
da es
(Hakala
e
al.,
2012).
The
da a
consis ed
o
a
se
o
112
mode n
cul i a s
o
bo h
Finnish
and
o eign
o igin
om
he
ea ly
1980s
o
he
p esen
(8.430
eco ds)
(Table
1).
The
ag o-clima ic
da a
o
he
Finnish
Me eo ological
Ins i u e
o
he
ial
loca ions
we e
used.
Ten
ag o-clima ic
pa ame e s
ha
mos
a ec ed
ba ley
g ain
yield
in
he
ials
we e
iden ified
using
a
eg ession
analysis
o
pa ame e s,
which
we e
selec ed
based
on
p e ious
li e a u e
and
obse a ions
( o
de ails,
see
Hakala
e
al.,
2012).
The
co ela ing
pa ame e s
we e
excluded
o
a oid
mul i-
collinea i y.
Two
addi ional
pa ame e s
(pa ame e s
9
and
10
below)
we e
selec ed
based
on
he
ecen
Eu opean
s udy
by
T nka
e
al.
(2011).
Consequen ly,
he
ollowing
wel e
phenology-
ela ed
ag o-clima ic
pa ame e s,
which
a e
he
mos
c i ical
o
ba ley
pe o mance
in
Finland,
we e
selec ed.
(1)
P ecipi a ion
du ing
one
mon h
be o e
sowing
(mm).
(2)
De ia ion
om
a
fixed
ea ly
sowing
da e
(d).
(3)
D ough
3–7
weeks
a e
sowing
indica ed
by
accumula ed
p ecipi a ion
(mm).
(4)
Hea
s ess
days
o
25
8C
one
week
be o e
h ough
wo
weeks
a e
heading
(d).
Fig.
1.
The
p oposed
app oach
o
esponse
di e si y
assessmen .
The
s eps
o
he
gene ic
p ocedu e
a e
p esen ed
in
bold.
The
p ocedu e
ha
is
applied
o
he
case
is
specified
o
each
s ep.
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
187
(5)
Ex eme
hea
s ess
days
o
28
8C
one
week
be o e
h ough
wo
weeks
a e
heading
(d).
(6)
Tempe a u e
sum
(T
sum
>
5
8C)
accumula ion
om
14
d
be o e
heading
un il
heading
(8C)
(T
sum
>
5
8C
is
he
sum
o
deg ees
abo e
5
8C
o
all
days,
o
which
T
mean
>
5
8C).
(7)
T
sum
>
5
8C
accumula ion
a e
om
heading
un il
yellow
ipeness
(8C).
(8)
T
sum
>
5
8C
accumula ion
a e
pe
day
om
heading
un il
yellow
ipeness
(8C).
(9)
Sum
o
e ec i e
global
adia ion
om
sowing
un il
yellow
ipeness
(MJ
m
2
o
days
wi h
T
mean
>
5
8C).
(10)
Sum
o
e ec i e
g owing
days
om
sowing
un il
yellow
ipeness
(d)
(numbe
o
days
wi h
T
mean
>
5
8C).
(11)
Numbe
o
days
wi h
ain
(>1
mm)
om
sowing
un il
yellow
ipeness
(d).
(12)
Seasonal
p ecipi a ion
om
sowing
un il
yellow
ipeness
(mm).
2.1.2.
S ep
2:
es ima ing
componen
esponses
o
he
ac o s
Each
ag o-clima ic
pa ame e
was
classified
in o
h ee
ca ego-
ies
because
he
ela ions
be ween
he
g ain
yield
and
he
ag o-
clima ic
pa ame e s
we e
nonlinea
in
mos
cases.
The
33 d
and
he
66 h
pe cen iles
we e
used
o
o m
equal-sized
ca ego ies.
Fo
example,
he
g ain
yield
obse a ions
o
each
ba ley
cul i a
we e
di ided
in o
g oups
based
on
p ecipi a ion
a es
o
0–24
mm,
be ween
24
and
40
mm
and
abo e
40
mm
one
mon h
be o e
Fig.
2.
Dispa i y
be ween
he
Shannon
indices
and
he
equi abili ies
o
he
ba ley
cul i a
ype
di e si y
(con inuous
line)
and
esponse
di e si y
(dashed
line).
Equi abili y
ep esen s
he
e enness
componen
o
he
di e si y
indices
which
also
include
he
componen
o
ichness.
Equi abili y
was
calcula ed
by
di iding
each
alue
o
he
Shannon
di e si y
index
by
he
heo e ical
maximum
o
ha
alue.
The
de elopmen
in
he
egions
wi h
he
smalles
and
g ea es
dispa i y
be ween
he
indices
since
2005
is
shown.
Da k
g een
indica es
he
egions
o
which
he
dispa i y
alues
we e
in
he
lowe
hal
o
all
egional
alues
( he
cha s
o
he
le ).
The
size
o
he
ci cles
illus a es
he
ba ley
cul i a ion
a ea
in
2005–2009.
The
Roman
nume als
e e
o
he
egions
ha
a e
p esen ed
in
Table
1.
Table
1
Cha ac e is ics
o
he
cul i a
da a.
The
15
ba ley
cul i a s
ha
we e
used
in
he
alida ion
a e
shown
as
examples.
Cul i a
Fi s
ial
Las
ial
Numbe
o
ials
Mean
yield
(kg
ha
1
)
STD
o
yield
Mean
hec oli e
weigh
(kg)
STD
o
hec oli e
weigh
Heading
DAS
a
Yellow
ipeness
DAS
a
A u i
1989
2008
133
4911
1316
64.0
4.3
53.8
95.1
A e
1987
2003
274
4727
1389
62.9
5.2
53.4
93.6
Ba ke
1997
2009
28
4702
1318
68.7
4.2
53.8
91.7
E kki
1992
2008
99
5386
1366
65.6
4.2
55.3
96.0
Ina i
1991
2006
67
4898
1586
69.4
3.8
53.1
93.6
Jy a
¨1997
2008
69
4729
1483
66.5
4.0
54.5
91.4
Kunna i
1997
2009
144
5089
1510
65.6
4.6
54.8
91.4
Kus aa
1981
2001
290
4221
1474
67.3
5.6
52.3
92.9
Kymppi
1981
1999
249
4507
1592
66.0
5.5
52.9
93.3
Lo iisa
1985
1996
173
4709
1530
64.1
6.7
52.2
94.4
Me e
1982
1997
122
4589
1450
66.6
6.5
52.7
94.0
Rolfi
1990
2009
179
4880
1380
62.7
4.6
54.8
95.0
Saana
1992
2008
124
4696
1327
67.3
4.5
55.1
92.2
Sca le
1995
2009
120
4690
1665
69.3
3.7
55.0
91.8
Thule
1991
1999
85
5050
1380
65.0
4.7
53.9
95.9
a
DAS,
days
a e
sowing.
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
188
sowing.
Howe e ,
he
ex eme
hea
s ess
days
o
28
8C
one
week
be o e
h ough
wo
weeks
a e
heading
( he
ag o-clima ic
pa ame e
(5)
abo e),
we e
dis ibu ed
among
he
62nd
(0
days)
and
he
72nd
(1
day
o
less)
pe cen iles,
while
he
es
o
he
cases
ep esen ed
mo e
han
1
hea
s ess
day
o
28
8C.
The
in e ac ion
o
hese
ca ego ies
wi h
he
cul i a
g ain
yield
o
each
o
he
112
mode n
cul i a s
(see
Sec ion
2.1.1)
was
analysed
using
he
ollowing
mixed
model:
y
i
jklm
¼
m
þ
cul i a
i
þ
ca ego y
j
þ
cul i a

ca ego y
i
j
þ
expe imen al
si e

yea

ialðca ego yÞ
klm
j
þ
e
i
jklm
whe e
y
ijklm
is
he
obse ed
yield,
m
is
he
in e cep ,
cul i a
i
is
he
a e age
yield
le el
o
i h
cul i a ,
ca ego y
j
is
he
a e age
yield
le el
a
j h
le el
o
ca ego ised
en i onmen
(j
=
1,
2,
3)
and
cul i a

ca ego y
ij
is
he
cul i a -by-en i onmen
in e ac ion.
All
he
abo e
e ec s
a e
fixed
in
he
model.
Expe imen al
si e

yea

ial(ca ego y)
klmj
is
he
andom
e ec
o
k h
expe imen al
si e,
l h
yea
and
m h
ial
wi hin
j h
ca ego y,
and
e
ijklm
is
a
no mally
dis ibu ed
esidual
e o .
The
cul i a -by-
en i onmen
in e ac ion
was
s a is ically
significan
(P
<
0.05)
o
e e y
ag o-clima ic
pa ame e
included.
Fo
each
cul i a
and
ag o-clima ic
pa ame e ,
he
di e ence
in
yield
be ween
he
ex eme
ca ego ies
1
and
3
was
calcula ed.
These
da a
consis ed
o
he
g ain
yield
esponses
o
112
cul i a s
o
12
ag o-clima ic
pa ame e s.
Fo
example,
he
mean
yield
o
each
cul i a
o
p ecipi a ion
a es
o e
40
mm
one
mon h
be o e
sowing
(ag o-clima ic
pa ame e
1,
ca ego y
3)
we e
sub ac ed
om
he
mean
yield
o
each
cul i a
o
p ecipi a ion
a es
below
24
mm
one
mon h
be o e
sowing
(ag o-clima ic
pa ame e
1,
ca ego y
1).
Consequen ly,
a
posi i e
g ain
yield
esponse
mean
ha
he
g ain
yield
was
be e
when
he
p ecipi a ion
a e
one
mon h
be o e
sowing
was
low.
2.1.3.
S ep
3:
alida ing
he
esponses
wi h
o he
da a
The
alidi y
o
he
es ima ed
yield
esponses
in
he
ials
was
es ed
unde
he
condi ions
occu ing
on
a ms,
o
ensu e
he
alidi y
o
he
conclusions
o
p ac ical
ag icul u e.
Da a
on
he
cul i a
g ain
yield
on
a ms,
which
we e
collec ed
by
he
Ce eal
Inspec ion
Uni
o
he
Finnish
Food
Sa e y
Au ho i y
since
1966,
and
he
g id-based
wea he
da a
a
a
10
km

10
km
esolu ion
o
he
Finnish
Me eo ological
Ins i u e,
which
o igina ed
om
he
p oxima e
wea he
s a ions,
we e
used.
In
o al,
1700
egionally
ep esen a i e
a ms
we e
moni o ed,
and
app oxima ely
one-
hi d
o
hese
a ms
we e
e-selec ed
annually
a
andom.
The
cul i a ion
p ac ices
and
yields
we e
documen ed
by
he
a me s,
and
he
hec oli e
g ain
weigh s
we e
assessed
in
he
labo a o y
om
samples
ha
we e
p o ided
by
he
a me s.
The
ag o-clima ic
pa ame e s
we e
adjus ed
o
he
phenological
s ages
by
modelling
he
c i ical
phenological
da es
(T nka
e
al.,
2011),
which
we e
based
on
he
sowing
da es
ha
we e
documen ed
on
each
a m.
The
in e ac ion
o
he
g ain
yield
o
each
cul i a
and
each
ag o-
clima ic
pa ame e
was
es ed
in
he
a m
da a
in
a
simila
manne
o
ha
o
he
ial
da a.
Pea son’s
co ela ion
coe ficien
was
calcula ed
o
compa e
he
ial
and
a m
da a
o
he
cul i a
g ain
yield
esponses
o
he
ag o-clima ic
pa ame e s.
To
con ol
he
possible
bias
ha
migh
ha e
been
in oduced
by
he
a me s’
yield
assessmen s,
he
co ela ion
o
he
ial
e sus
a m
da a
o
he
hec oli e
weigh s
ha
we e
assessed
in
he
labo a o y
was
also
calcula ed.
The
hec oli e
weigh s
we e
only
used
o
his
pu pose.
Due
o
he
po en ially
high
a ia ion
in
a m
condi ions,
which
may
a ec
cul i a
esponses
o
he
ag o-
clima ic
pa ame e s,
only
he
cul i a s
o
which
he e
we e
mo e
han
100
a m
obse a ions
(15
cul i a s)
we e
selec ed
o
alida ion.
The
da a
om
1998
o
2005
we e
used
o
alida ion,
he
pe iod
being
limi ed
by
he
a ailabili y
o
g id-based
ag o-clima ic
pa ame e s
( adia ion).
In
addi ion
o
he
co ela-
ions,
also
a
p incipal
componen
analysis
o
bo h
he
ial
and
he
a m
da a
was
pe o med,
o
compa e
he
ial
and
a m
esul s
o
alida ion.
The
esul s
a e
en a i e
due
o
he
ela i ely
low
numbe
o
analysed
uni s
ela i e
o
he
equi emen s
o
a
obus
p incipal
componen
analysis:
40
cul i a s
in
he
a m
da a
we e
used.
2.2.
S age
II:
es ima ion
o
esponse
di e si y
S age
II
classifies
he
componen s
acco ding
o
he
esponses
and
c ea es
a
di e si y
index,
which
is
based
on
he
classifica ion.
2.2.1.
S ep
4:
cons uc ing
he
esponse
di e si y
index
A
clus e
analysis
using
Wa d’s
me hod
(Wa d,
1963)
was
employed
o
he
da a
ha
we e
c ea ed
in
S ep
2
o
clus e
he
cul i a s
acco ding
o
g ain
yield
esponses
o
he
ag o-clima ic
pa ame e s.
The
clus e ing
was
based
on
a
Mahalanobis
dis ance
ma ix,
which
uses
he
ull
mul i a ia e
in o ma ion
o
he
g ain
yield
esponses
(McLachlan,
1999).
The
da a
con ained
he
g ain
yield
esponses
o
he
112
cul i a s
( ows)
o
he
12
ag o-clima ic
a iables
(columns).
The
Mahalanobis
dis ance
gi es
less
weigh
o
a iables
wi h
a
high
a iance
and
o
highly
co ela ed
a iables,
such
ha
all
he
cha ac e is ics
a e
ea ed
as
being
equally
impo an
(Mimmack
e
al.,
2001).
The
clus e
numbe
was
selec ed
based
on
he
dend og am,
he
pseudo
2
-c i e ion
and
he
-squa e
(Yeo
and
T uxillo,
2005)
a ia ion.
The
Shannon
di e si y
index
(H),
which
implies
bo h
ichness
and
e enness
o
dis ibu ion
(Shannon
and
Wea e ,
1949),
was
calcula ed
o
he
cul i a ion
a eas
o
he
12
clus e s
o
ba ley
cul i a s,
which
esul ed
om
clus e ing
(see
abo e)
in
he
16
adminis a i e
egions
o
Finland.
The
‘ esponse
di e si y’
index
hus
had
each
o
he
12
clus e s
as
a
di e si y
uni .
The
Shannon
di e si y
index
was
calcula ed
acco ding
o
he
ollowing
equa ion:
H
i
¼
X
K
k¼1
w
ik
W
i
ln w
ik
W
i
;
o
i
¼
1;
.
.
.
;
n
egions
whe e
k
=
1,
.
.
.,
K
e e s
o
he
numbe
o
clus e s;
w
ik
is
he
sum
o
cul i a ion
a ea
(ha)
by
clus e
k
o
egion
i,
W
i
ep esen s
he
o al
sum
o
cul i a ion
a ea
(ha)
o
egion
i,
and
ðw
ik
=W
i
Þ
is
he
p opo ion
o
he
cul i a ion
a ea
(ha)
ha
is
co e ed
by
clus e
k.
Shannon’s
equi abili y
o
he
annual
cul i a ion
a ea
o
he
clus e s
was
also
calcula ed
o
illus a e
independen ly
he
e enness
componen
o
he
di e si y
index
(Mulde
e
al.,
2004).
The
equi abili ies
also
allow
a
di ec
compa ison
o
he
shi s
in
ype
di e si y
e sus
in
esponse
di e si y
because
he
scale
o
each
is
he
same
o
equi abili y.
Shannon’s
equi abili y
(E
H
)
was
calcula ed
by
di iding
each
H
alue
by
i s
heo e ical
maximum
(H
max
):
(H
max
=
ln(K)).
The
possible
equi abili y
alues
ange
be ween
0
and
1,
wi h
1
indica ing
comple e
e enness.
The
cul i a ion
a eas
o
he
ba ley
cul i a s
o
1998–2009
we e
used
o
calcula e
he
di e si y
indices
and
equi abili ies.
This
in o ma-
ion
was
collec ed
annually
om
all
a ms
in
Finland
by
he
In o ma ion
Cen e
o
he
Minis y
o
Ag icul u e
and
Fo es y.
2.2.2.
S ep
5:
assessing
he
alue
added
by
esponse
di e si y
The
annual
Shannon
di e si y
index
and
he
equi abili y
o
each
o
he
16
egions
we e
calcula ed
using
each
indi idual
cul i a
as
a
di e si y
uni
(‘ ype
di e si y’)
(Himanen
e
al.,
2013a)
o
compa ison
wi h
he
‘ esponse
di e si y’
index
ha
was
cons uc ed
(Sec ion
2.2.1,
S ep
4).
Di e ences
be ween
he
slopes,
which
illus a ed
he
de elopmen
o
he
di e si y
indices
and
equi abili ies
( o
‘ esponse
di e si y’
and
‘ ype
di e si y’)
o e
ime,
we e
es ed
o
bo h
indices
and
equi abili ies.
The
slopes
o
he
a iable
yea
o
he
indices
and
he
equi abili ies
we e
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
189
calcula ed
o
each
egion
using
a
linea
eg ession
model.
The
models
consis ed
o
he
in e cep
e m
in
addi ion
o
he
yea .
The
equali y
o
he
slopes
wi hin
each
egion
was
es ed
o
he
indices
and
equi abili ies
using
S uden ’s
wo- ailed
- es .
Fo
he
equi abili ies
and
indices,
he
di e ence
o
he
annual
means
o
he
wo
indices
was
also
es ed
using
he
ollowing
mixed
model:
y
i
jk
¼
m
þ
index
i
þ
egion
j
þ
index

egion
i
j
þ
yea

egion
k
j
þ
e
i
jk
whe e
y
ijk
is
he
obse ed
alue
o
index,
m
is
he
in e cep ,
index
i
is
he
a e age
le el
o
i h
index,
egion
j
is
he
a e age
le el
a
he
j h
le el
o
index
(i
=
1,
2)
and
index

egion
ij
is
he
in e ac ion
o
he
i h
index
wi hin
he
j h
egion.
The
index
e e s
o
bo h
H
and
E
H
.
All
he
abo e
e ec s
a e
fixed
in
he
model.
Yea

egion
kj
is
he
andom
e ec
o
he
k h
yea
wi hin
he
j h
egion,
and
e
ijk
is
he
no mally
dis ibu ed
esidual
e o .
All
o
he
s a is ical
analyses
we e
pe o med
using
PROC
MIXED,
CORR,
FACTOR,
PRINCOMP,
DISTANCE,
CLUSTER
and
REG
o
SAS
( e sion
9.3,
SAS
Ins i u e
Inc.,
Ca y,
NC,
USA).
PROC
PRINCOMP
and
DISTANCE
we e
used
o
calcula e
he
Mahalanobis
ma ix
in
S ep
4
(Sec ion
2.2.1).
3.
Resul s
A
gene ic
p ocedu e
is
p oposed
and
he
alue-added
o
he
esponse
di e si y
app oach
is
demons a ed
by
exempli ying
he
p ocedu e
using
he
case
o
ba ley
cul i a s
(Fig.
1).
The
p ac ical
significance
o
he
yield
esponse
o
he
ag o-
clima ic
pa ame e s
is
illus a ed
by
he
di e ence
in
cul i a
yield
be ween
he
highes
and
lowes
hi d
o
he
alues
o
he
12
pa ame e s.
The
median
o
he
cul i a s
in
such
di e ences
anged
be ween
134
and
579
kg
ha
1
(332
kg
ha
1
on
a e age)
depending
on
he
pa ame e .
Yield
esponses
o
up
o
1500
kg
ha
1
o
some
pa ame e s
we e
demons a ed,
and
a
esponse
o
mo e
han
1000
kg
ha
1
was
no
a e.
A
clus e
numbe
o
12
o
he
g ain
yield
esponses
by
he
cul i a s
o
he
12
ag o-clima ic
pa ame e s
was
iden ified
as
bes
co esponding
o
he
s a is ical
c i e ia
( o
he
c i e ia,
see
Sec ion
2.2.1).
The
p opo ion
o
a ia ion
in
yield
esponses
o
wea he
explained
by
he
12
clus e s
was
0.43.
The
co ela ion
be ween
he
cul i a
esponses
in
he
ial
da a
and
a m
da a
was
0.58
[CI
95%
0.48,
0.67]
o
yield
and
0.70
[CI
95%
0.61,
0.76]
o
hec oli e
weigh .
The
p incipal
componen
analysis
o
he
g ain
yields
esul ed
in
a
simila
p incipal
componen
s uc u e
o
bo h
he
ial
da a
and
he
a m
da a.
The e
we e
se e al
cul i a s
in
cul i a ion
o
mos
o
he
ag o-
clima ic
esponse
clus e s
ep esen ed
in
he
egions.
The e o e,
he
means
o
he
cul i a
ype
di e si y
indices
we e
highe
han
esponse
di e si y
indices
o
he
sown
a eas
o
ba ley
cul i a s
(P
<
0.0001)
(Fig.
2).
In
he
sou he n
egions,
bo h
o
he
indices
inc eased
e enly
in
alue
om
1998
o
2009
(Fig.
2
and
Table
2).
Howe e ,
in
mo e
han
hal
o
he
16
egions
o
he
coun y,
i.e.
in
he
cen al
and
no he n
egions,
he
slope
o
he
esponse
di e si y
index
di e ed
om
ha
o
he
ype
di e si y
index
(Table
2).
In
he
cen al
and
no he n
egions,
he
esponse
di e si y
index
dec eased,
al hough
he
cul i a
ype
di e si y
index
con inuously
inc eased.
The
disc epancy
be ween
he
wo
indices
in
he
cen al
and
no he n
egions
ended
o
inc ease
sligh ly
a
he
s a
o
he
2000s
and
inc eased
again
in
he
middle
o
he
decade
(Fig.
2).
The
dec ease
o
he
equi abili ies
(e enness)
o
esponse
di e si y
was
highe
han
o
he
esponse
di e si y
index
as
a
whole
(Fig.
2)
showing
he
ba ley
cul i a ion
concen a ing
in
ewe
ag o-clima ic
esponse
clus e s,
while
he
numbe
o
esponse
clus e s
ep esen ed
( ichness)
inc eased
li le
ela i e
o
he
inc ease
in
he
numbe
o
cul i a s.
The
dec ease
in
he
equi abili y
o
ba ley
cul i a
esponse
di e si y
in
he
Cen al
and
No he n
Finland
coincided
wi h
he
inc ease
in
he
cul i a ion
a ea
o
a
single
esponse
clus e ,
he
g ain
yield
o
which
is
educed
by
d ough
and
which
benefi s
om
a
ela i ely
ea ly
sowing.
This
clus e
(Clus e
3)
eplaced
cul i a s
om
ano he ,
p e iously
equally
ex ensi ely
cul i a ed
esponse
clus e
(Clus e
1)
(Fig.
3),
which
shows
li le
esponse
o
he
wea he
pa ame e s
bu
wi h
only
a
mode a e
yield
le el.
4.
Discussion
4.1.
Value-added
by
empi ical
assessmen
o
esponse
di e si y
The
esilience
app oach
is
a
pe spec i e
o
o ien a ion
in
unce ain y,
complexi y
and
unp edic able
a ia ion,
sugges ing
adap i e
managemen .
The
p oposed
p ocedu e
assis s
adap i e
Table
2
Di e ences
be ween
cul i a
ype
and
esponse
di e si y
indices
and
hei
equi abili ies.
Region
a
Di e si y
indices
Equi abili ies
b
Di e ence
in
slopes
c
p
alue
d
Di e ence
in
slopes
c
p
alue
d
Di e ence
in
means
c
p
alue
d
I
0.07

0.02
<0.001
0.02

0.01
0.024
0.13

0.02
<0.001
II
0.12

0.01
<0.001
0.04

0.00
<0.001
0.17

0.02
<0.001
III
0.11

0.01
<0.001
0.04

0.01
<0.001
0.03

0.02
0.207
IV
0.05

0.01
<0.001
0.01

0.00
0.029
0.19

0.02
<0.001
V
0.03

0.01
<0.001
0.00

0.00
0.524
0.15

0.02
<0.001
VI
0.08

0.01
<0.001
0.02

0.00
<0.001
0.21

0.02
<0.001
VII
0.08

0.01
<0.001
0.03

0.01
<0.001
0.17

0.02
<0.001
VIII
0.08

0.01
<0.001
0.03

0.00
<0.001
0.20

0.02
<0.001
IX
0.01

0.01
0.204
0.00

0.00
0.641
0.11

0.02
<0.001
X
0.01

0.01
0.339
0.01

0.01
0.069
0.08

0.02
0.002
XI
0.02

0.01
0.048
0.01

0.00
<0.001
0.01

0.02
0.725
XII
0.02

0.01
0.065
0.01

0.00
0.264
0.07

0.02
0.005
XIII
0.01

0.01
0.058
0.01

0.00
0.129
0.07

0.02
0.006
XIV
0.02

0.01
0.028
0.02

0.00
<0.001
0.04

0.02
0.150
XV
0.00

0.01
0.904
0.02

0.00
<0.001
0.00

0.02
0.875
XVI
0.02

0.01
0.115
0.01

0.01
0.351
0.05

0.02
0.038
a
The
oman
numbe s
e e
o
he
egions
in
Fig.
2.
b
Each
alue
o
Shannon
di e si y
index
di ided
by
he
heo e ical
maximum
o
ha
alue,
ep esen ing
e enness.
c
The
di e ence
be ween
he
indices
(index
ype

index
esponse
)

s anda d
e o
o
di e ence.
The
slope
e e s
o
he
a e age
annual
change
in
he
alue
o
he
index
and
he
mean
e e s
o
he
a e age
alue
o
he
index
1998–2009.
d
S uden ’s
wo- ailed
- es ,
a
=
0.05,
n
=
24.
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
190

managemen
h ough
he
empi ical
assessmen
o
he
c i ical
ac o s
o
change,
and
h ough
iden ifica ion
o
he
di e si y
mos
e ec i e
o
educing
sensi i i y
o
a ia ion
and
inc easing
he
capaci y
o
adap
o
plausible
anges
o
such
c i ical
ac o s.
The e o e,
he
p ocedu e
has
ele an
implica ions
o
public
policies
and
p i a e
en e p ise
s a egies.
The
p ocedu e
p o ides
means
o
communica e
a
he
in e aces
o
science,
policy
and
p ac i ione s,
and
o
acili a e
public-p i a e
pa ne ships.
The
gene ic
app oach
p oposed
he e
can
guide
adap i e
managemen
and
go e nance
no
only
in
he
case
o
clima e
a iabili y
and
change
such
as
demons a ed
he e
bu
also
in
he
esponse
by
he
economy
o
a m
ac i i ies
o
by
sales
o
e ail
supplie s
o
p ice
ola ili y
(Howden
e
al.,
2007),
o
example.
In
he
la e
case,
which
exemplifies
he
on-going
wo k
o
pa
o
he
au ho s,
ood
supplie s
could
be
clus e ed
acco ding
o
he
di e en ial
esponses
o
hei
sales
o
global
p ice
a iabili y.
High
clus e
di e si y
would
indica e
s abili y
ega ding
consume
access
o
a o dable
ood.
The
dec ease
in
esponse
di e si y
o
ba ley
cul i a s
in
he
cen al
and
no he n
egions
o
Finland
shown
he e,
indica es
inc eased
ulne abili y
and
dec eased
esilience
(Folke
e
al.,
2004;
Elmq is
e
al.,
2003;
Lalibe e
e
al.,
2010),
which
was
no
e ealed
me ely
by
he
cul i a
di e si y
(‘ ype
di e si y’).
The
de elop-
men s
ha
a e
dele e ious
o
esilience,
as
e ealed
by
he
empi ical
assessmen
o
he
esponse
di e si y,
can
hen
be
add essed
h ough
adap i e
managemen
in o med
by
he
assessmen .
Simila
models,
as
he e
o
ba ley,
can
be
cons uc ed
o
o he
c ops
and
condi ions
based
on
documen ed
plu i-annual
yields
and
associa ed
wea he .
Such
models
would
acili a e
a ge ed
c op
di e sifica ion
beyond
main aining
biodi e si y,
o
se e
as
a
ool
o
a me s
o
enhance
esilience
and
adap i e
capaci y
(Ja is
e
al.,
2008).
Simila
models
can
be
used
o
quan i y
esponse
di e si y
gene ally.
The
demons a ed
app oach
is
applicable
o
any
sys em
in
which
empi ical
in o ma ion
o
he
esponse
o
componen s
can
be
ela ed
o
documen ed
changes
and
a ia ions
wi h
ele ance.
The
added
alue
o
he
use
o
his
p oposed
app oach
can
be
in es iga ed
in
each
case
by
compa ing
he
di e si y
index
o
he
esponses
wi h
he
index
o
he
me e
ypes.
The
limi ing
ac o
o
he
applica ion
o
he
p oposed
app oach
could
in
many
cases
be
se
by
he
a ailabili y
o
da a.
Conce ning
na u al
and
managed
ecosys ems,
he
da a
equi emen s
a e
me
by
c ea ing
long- e m
obse a o ies
and
well-planned
moni o ing
in as uc u es
ha
p o ide
eliable
mul i-yea
da ase s.
Examples
o
such
can
be
ound
in
Eu ope
h ough
he
ANAEE
ne wo king
ini ia i e
(www.anaee.com)
o ,
mo e
specifically
o
g asslands
and
o es s,
h ough
he
Ecofinde s
p ojec
(www.ecofinde s.eu).
Ch onosequences
may
also
be
sou ces
o
da a
o
hese
esponse
di e si y
assessmen s.
Howe e ,
many
mo e
ypes
o
da a
se s
can
be
u ilised
in
applying
he
p oposed
p ocedu e
in
a ious
con ex s.
Examples
o
such
da a
se s
include
he
Eu opean
Fa m
Accoun-
ancy
Da a
Ne wo k
(h p://ec.eu opa.eu/ag icul u e/ ica/defini-
ions_en.c m),
nume ous
o he
da a
se s
ha
ha e
been
compiled
by
au ho i ies,
and
da a
se s
by
e aile s
and
o he
p i a e
ac o s.
In
addi ion
o
hose
me hods
applied
in
his
s udy,
he e
a e
o he
me hods
ha
can
also
be
applied,
depending
on
he
con ex
o
he
applica ion.
P incipal
componen
analysis
would
be
an
al e na i e,
in
addi ion
o
di ec
clus e ing,
o
model
he
esponse
s uc u e
and
in
alida ion
o
ensu e
applicabili y
in
he
con ex
whe e
he
model
could
se e
decision-making.
4.2.
Value-added
by
he
assessmen
o
esponse
di e si y
o
ba ley
cul i a s
in
Finland
The
unce ain y
in
clima e
change
is
g ea es
a
he
local
le el
whe e
indi idual
a me s
ope a e
(Howden
e
al.,
2007;
Ro
¨ e
e
al.,
2013).
The
a me s
manage
c op
cul i a
di e si y
annually.
The
pa icula
clus e -based
esponse
di e si y
index
o
ba ley
cul i a s
o
wea he
is
di ec ly
applicable
o
a ms
in
Finland.
The
s a is ically
significan
and
ela i ely
high
posi i e
co ela ions
be ween
he
cul i a
esponses
o
he
ag o-clima ic
pa ame e s
in
he
ial
da a
e sus
in
he
da a
om
a ms
show
he
alidi y
o
he
esponse
di e si y
index
o
p ac ical
ag icul u e,
despi e
po en-
ially
mo e
a ia ion
in
condi ions
and
less
p ecise
wea he
es ima es
on
a ms
han
in
he
ials.
The
ac
ha
he
co ela ion
coe ficien
o
hec oli e
weigh s
o
he
ial
da a
e sus
a m
da a
did
no
essen ially
di e
om
he
co esponding
co ela ion
coe ficien
o
g ain
yields
indica es
ha
he
a me
assessmen s
o
he
g ain
yields
we e
eliable
enough.
The
simila
p incipal
componen
s uc u e
ound
o
bo h
he
ial
and
he
a m
da a
p o ides
an
addi ional
e idence
o
he
conclusion
ha
he
esponse
di e si y
model
ha
was
cons uc ed
is
applicable
unde
a m
condi ions.
The
dec ease
in
he
esponse
di e si y
o
ba ley
cul i a s
in
cen al
and
no he n
Finland
du ing
he
las
decade,
despi e
he
con inuous
inc ease
in
cul i a
( ype)
di e si y,
was
due
o
he
cul i a ion
a ea
concen a ing
on
ewe
wea he
esponse
clus e s
o
ba ley
cul i a s
especially
a
he
la e
hal
o
he
decade.
One
Fig.
3.
De elopmen
o
he
cul i a ion
a ea
o
he
ba ley
cul i a
esponse
clus e s
(1998–2009)
in
he
Sou he n
egions
(le )
and
in
he
Cen al
and
No he n
egions
( igh )
o
Finland.
Clus e
3
ep esen s
cul i a s,
he
g ain
yield
o
which
is
clea ly
educed
by
d ough
and
benefi s
om
ela i ely
ea ly
sowing.
Clus e
1
ep esen s
cul i a s
wi h
a
s able
bu
only
mode a e
yield.
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
191
wea he
esponse
clus e ,
wi h
cul i a s
sensi i e
o
d ough
and
benefi ing
om
ea ly
sowing,
inc easingly
domina ed.
The e
occu ed
no
shi ,
nei he
a
di e ence
be ween
sou he n
s.
cen al
and
no he n
Finnish
egions,
in
p ecipi a ion
o
empe a u e
du ing
he
g owing
seasons
o
1998–2009
(Himanen
e
al.,
2013a,b),
no
in
he
o he
ag o-clima ic
pa ame e s
ha
could
explain
he
obse ed
concen a ion
as
a
a me s’
coping
o
a
shi
in
wea he .
Ra he ,
i
seems
ha
a me s’
cul i a ion
concen a ed,
because
all
he
ba ley
cul i a s
pe o ming
well
in
hese
egions
in oduced
o
he
ma ke
du ing
he
pe iod
ep esen ed
he
same
wea he
esponse
clus e .
Nea ly
hal
o
all
he
ba ley
cul i a s
in oduced
o
he
ma ke
in
Finland
du ing
ha
pe iod
(1998–
2009),
and
65%
o
hei
accumula ed
cul i a ion
a ea,
89%
since
2005,
ep esen ed
ha
single
wea he
esponse
clus e
om
all
he
12
wea he
esponse
clus e s
o
ba ley
cul i a s
in
ials
du ing
he
las
decades.
On
he
con a y,
un il
1998,
ba ley
cul i a ion
a ea
was
mainly
di ided
among
wo
o
h ee
wea he
esponse
clus e s
wi h
a
dominan
one
di e en
om
ha
du ing
1998–2009.
The
inc eased
compe i ion
in
he
cul i a
ma ke
may
ha e
led
b eede s
o
elease
new
cul i a s
o
inc easing
simila i y.
In
cen al
and
no he n
Finland
ba ley
cul i a ion
(e en
i
ba ley
ep esen s
a
compa a i e
ad an age
in
ce eal
cul i a ion
in
no he n
Finland)
occu s
a
he
no he nmos
ma gin
o
global
ag icul u e,
wi h
a
ela i ely
na ow
gene ic
basis
o
use ul
b eeding
ma e ial.
The e o e
p ofi -o ien ed
b eeding
e o s
in
a
compe i i e
ma ke
whe e
new
cul i a s
always
ca ch
a en ion,
easily
concen a e
on
a
small
cul i a
g oup
o e ing
high
yield.
Fa me
expe imen a ion
may
hen
lead
o
an
inc easing
simila i y
among
sown
cul i a s,
unless
special
a en ion
is
gi en
and
ools
and
incen i es
o
inc ease
esponse
di e si y
a e
p o ided
o
p epa ing
o
a
clima e
wi h
high
unce ain y
(Ro
¨ e
e
al.,
2013)
and
inc eased
a iabili y
(Field
e
al.,
2012).
The
assessmen
ha
was
p oposed
he e
can
be
used
o
selec
a
ailo ed
se
o
cul i a s
ha
ep esen s
a
wide
ange
o
esponses
o
c i ical
wea he
a ia ion,
o
educe
he
in e -annual
a ia ion
and
p obabili y
o
yield
losses
on
a m,
in
a
pa icula
egion
and
o e
he
en i e
coun y.
The e o e,
he
model
was
alida ed
unde
a me
field
condi ions,
whe e
managemen
(e.g.
e ilisa ion,
c op
p o ec ion)
and
soil
ypes
a y
o
a
g ea e
ex en
and
he e o e
po en ially
elici
di e ences
in
esponse
o
wea he
in
compa ison
wi h
he
case
a
o ficial
ial
si es.
The
specific
model
could
be
used
in
he
communica ion
among
a me s,
ad iso s
and
b eede s,
and
o he
ac o s
such
as
indus y
and
ade
while
making
cul i a ion
con ac s.
The
use
o
such
models
could
be
p omo ed
by
adminis a o s
and
policy-make s
and,
o
ins ance,
h ough
he
Common
Ag icul u al
Policy
o
he
Eu opean
Union.
The
empi ical
assessmen
o
esponse
di e si y
could
also
se e
he
main enance
o
a
su ficien ly
b oad
ange
o
esponses
o
c i ical
wea he
in
b eeding,
o
secu e
he
adap i e
capaci y
o
he
long
e m
equi emen s.
The
assessmen
can
be
used
in
communica ing
among
p i a e
b eeding
companies,
au ho i ies
and
policy-make s
in
o de
o
sha e
he
cos s
o
such
a
public
good.
P ac ical
ools
applying
he
esul s
a e
unde
de elop-
men
o
assis
he
ac o s
in
communica ion
and
decision-making.
Ba ley
is
he
mos
widely
g own
ce eal
and
odde
c op
in
Finland,
and
i
is
di ficul
o
find
sui able
subs i u es
o
his
c op.
Thus,
he
dec easing
esilience
o
ba ley
cul i a ion
could
lead
o
a
decline
in
animal
p oduc ion
as
well.
Such
a
de elopmen
would
endange
ood
p ocessing
(dai y,
mea
and
b ewing
indus ies),
which
elies
on
domes ic
p ima y
p oduc ion.
Reduced
esilience
and
he
consequen
decline
in
ba ley
cul i a ion
when
acing
anomalies
in
c i ical
wea he
could
pu
many
ac i i ies
ha
suppo
Finnish
ag icul u e
a
isk,
such
as
b eeding,
educa ion,
ex ension,
seed
and
odde
ade
and
quali y
con ol
se ices.
I
c i ical
h esholds,
in
e ms
o
he
ex en
o
he
cu en ly
ela i ely
small
ma ke
o
such
p oduc s
and
se ices,
we e
eached,
a
domino
e ec
in
he
domes ic
ood
supply
chain
could
esul
ha
would
endange
Finland’s
ood
secu i y.
Such
a
h eshold
could
be
c ossed
due
o
one
o
se e al
yea s
o
los
ha es
and
he
consequen
need
o
ely
solely
on
expensi e
impo ed
odde .
Co espondingly,
mo e
di e si y
in
esponses
o
c i ical,
unp edic able
change
and
a ia ion
could
ensu e
ha
a
highe
deg ee
o
a ia ion
in
wea he
is
equi ed
be o e
a
c i ical
h eshold
in
he
ba ley
p oduc ion
sys em
and
ood
secu i y
would
be
c ossed.
This
example
illus a es
he
po en ial
o
e ealing
esponse
di e si y
in
dis ancing
c i ical
h esholds.
Resilience
can,
such
as
shown
he e,
only
be
enhanced
h ough
di e sifica ion
i
he
e y
aspec
o
esponse
di e si y
is
di ec ly
assessed
and
i
he
p ac ical
managemen
o
esilience
is
unde s ood
and
acili a ed
h ough
such
assessmen s.
5.
Conclusions
P ac ical
ools,
such
as
he
assessmen s
sugges ed
by
his
s udy,
could
p omo e
he
obus
oo ing
o
he
esilience
discou se
on
empi ical
g ounds,
an
on-going
conce n
in
he
esilience
commu-
ni y
(Folke
e
al.,
2004)
and
in
adap a ion
science
(Howden
e
al.,
2007).
Fo
he
equi ed
ans o ma ions
and
adap i e
esponses,
a
desi ed
adap i e
p ocess
a he
han
a
p ecisely
planned
ou come
is
sough
(e.g.
Milly
e
al.,
2008)
ha
se s
specific
demands
on
he
assessmen
app oaches
and
long- e m
moni o ing
sys ems,
which
a e
exemplified
he e
by
he
pa icula
case
o
Finnish
ba ley.
Such
p ac ical
ools
and
a ailable
da a
could
p e en
he
concep
o
esilience,
which
has
po en ial
o
open
new
pe spec i es,
om
simply
becoming
ano he
buzzwo d
among
many.
The
p oposed
gene ic
app oach
o
he
empi ical
iden ifica ion
o
esponse
di e si y
o
manage
esilience
and
adap i e
capaci y
o
global
en i onmen al
change
c ea es
added
alue
by
guiding
ailo ed
di e sifica ion.
I
he
key
di e si y
ha
os e s
esilience
is
iden ified,
mo e
esilience
can
be
achie ed
wi h
less
di e si y.
An
inc ease
in
he
e ficiency
o
di e sifica ion
would
help
o
success ully
combine
he
complemen a y
dimensions
o
sus ain-
abili y,
i.e.
esilience
and
e ficiency.
Acknowledgemen s
This
s udy
was
suppo ed
by
he
Clima e
Change
Adap a ion
P og am
ISTO
p ojec
ADACAPA,
and
by
he
Academy
o
Finland,
he
p ojec s
A-La-Ca e
(decision
no
140870)
and
ADIOSO
(decision
no
255954).
The
so wa e
de elopmen
o
assessing
he
ag o-
clima ic
condi ions
was
unded
by
he
Czech
Agency
o
Ag icul u al
Resea ch,
P ojec
No.
QI91C054,
and
he
applica ion
o
he
so wa e
was
unded
by
a
P ojec
De eloping
A
Mul idisci-
plina y
Scien ific
Team
ha
was
ocused
on
d ough
(No.
CZ.1.07/
2.3.00/20.0248).
Ou i
Ma
¨kila
¨p o ided
echnical
suppo
in
d awing
he
figu es.
Re e ences
Aubin,
I.,
Ouelle e,
M.H.,
Legend e,
P.,
Messie ,
C.,
Boucha d,
A.,
2009.
Compa ison
o
wo
plan
unc ional
app oaches
o
e alua e
na u al
es o a ion
along
an
old-
field
–
deciduous
o es
ch onosequence.
J.
Veg.
Sci.
20,
185–196.
Chapin,
F.S.,
e
al.,
1997.
Bio ic
con ol
o e
he
unc ioning
o
ecosys ems.
Science
277,
500–504.
Coumou,
D.,
Rahms o ,
S.,
2012.
A
decade
o
wea he
ex emes.
Na .
Clim.
Change
2,
491–496.
Dessai,
S.,
O’B ien,
K.,
Hulme,
M.,
2007.
Edi o ial:
on
unce ain y
and
clima e
change.
Global
En i on.
Change
17,
1–3.
Diaz,
S.,
Syms ad,
A.J.,
Chapin,
F.S.,
Wa dle,
D.A.,
Huenneke,
L.F.,
2007.
Inco po a ing
plan
unc ional
di e si y
e ec s
in
ecosys em
se ice
assessmen s.
P oc.
Na l.
Acad.
Sci.
U.S.A.
104,
20684–20689.
Elmq is ,
T.,
Folke,
C.,
Nys o
¨m,
M.,
Pe e son,
G.,
Beng sson,
J.,
Walke ,
B.,
No be g,
J.,
2003.
Response
di e si y
and
ecosys em
esilience.
F on .
Ecol.
En i on.
1,
488–494.
Field,
C.B.,
e ,
al.
(Eds.),
2012.
Managing
he
Risks
o
Ex eme
E en s
and
Disas e s
o
Ad ance
Clima e
Change
Adap a ion:
Summa y
o
Policymake s.
A
Special
Repo
o
Wo king
G oups
I
and
II
o
he
In e go e nmen al
Panel
on
Clima e
Change.
.
p.
19.
A ailable
om:
h p://ipcc-wg2.go /SREX/images/uploads/
SREX-SPMb ochu e_FINAL.pd .
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
192
Folke,
C.,
Ca pen e ,
S.,
Walke ,
B.,
Sche e ,
M.,
Elmq is ,
T.,
Gunde son,
L.,
Holling,
C.S.,
2004.
Regime
shi s,
esilience,
and
biodi e si y
in
ecosys em
manage-
men .
Annu.
Re .
Ecol.
E ol.
Sci.
35,
557–581.
Hakala,
K.,
Jauhiainen,
L.,
Himanen,
S.J.,
Ro
¨ e ,
R.P.,
Salo,
T.,
Kahiluo o,
H.,
2012.
Sensi i i y
o
ba ley
a ie ies
o
wea he
in
Finland.
J.
Ag ic.
Sci.
150,
145–160.
Himanen,
S.J.,
Ke oja,
E.,
Hakala,
K.,
Ro
¨ e ,
R.P.,
Salo,
T.,
Kahiluo o,
H.,
2013a.
Cul i a
di e si y
has
g ea
po en ial
o
inc ease
yield
o
eed
ba ley.
Ag on.
Sus ain.
De .
33,
519–530.
Himanen,
S.J.,
Hakala,
K.,
Kahiluo o,
H.,
2013b.
C op
esponses
o
clima e
and
socio-
economic
change
in
no he n
egions.
Reg.
En i on.
Change
13,
17–32.
Howden,
S.M.,
Soussana,
J.-F.,
Tubiello,
F.N.,
Chhe i,
N.,
Dunlop,
M.,
Meinke,
H.,
2007.
Adap ing
ag icul u e
o
clima e
change.
P oc.
Na l.
Acad.
Sci.
U.S.A.
104,
19691–19696.
Ja is,
D.I.,
B own,
A.H.D.,
Cuong,
P.H.,
Collado-Pandu o,
L.,
La ou ne ie-Mo eno,
L.,
Gyawali,
S.,
Tan o,
T.,
Sawadogo,
M.,
Ma ,
I.,
Sadiki,
M.,
Hue,
N.T.-N.,
A ias-Reyes,
L.,
Balma,
D.,
Baj acha ya,
J.,
Cas illo,
F.,
Rijal,
D.,
Belqadi,
L.,
Rana,
R.,
Saidi,
S.,
Oued aogo,
J.,
Zang e,
R.,
Rh ib,
K.,
Cha ez,
J.L.,
Schoen,
D.,
S hapi ,
B.,
De
San is,
P.,
Fadda,
C.,
Toby,
H.,
2008.
A
global
pe spec i e
o
he
ichness
and
e enness
o
adi ional
c op- a ie y
di e si y
main ained
by
a ming
communi ies.
P oc.
Na l.
Acad.
Sci.
U.S.A.
105,
5326–5331.
Lalibe e,
E.,
Wells,
J.A.,
DeCle ck,
F.,
Me cal e,
D.J.,
Ca e all,
C.P.,
Quei oz,
C.,
Aubin,
I.,
Bonse ,
S.P.,
Ding,
Y.,
F a e igo,
J.M.,
McNama a,
S.,
Mo gan,
J.W.,
Sanchez
Me los,
D.,
Vesk,
P.A.,
Mayfield,
M.M.,
2010.
Land-use
in ensifica ion
educes
unc ional
edundancy
and
esponse
di e si y
in
plan
communi ies.
Ecol.
Le .
13,
76–86.
McLachlan,
G.,
1999.
Mahalanobis
dis ance.
Resonance
4,
20–26.
Mimmack,
G.M.,
Mason,
S.J.,
Galpin,
J.S.,
2001.
Choice
o
dis ance
ma ices
in
clus e
analysis:
defining
egions.
J.
Clima e
14,
2790–2797.
Milly,
P.C.D.,
Be ancou ,
J.,
Falkenma k,
M.,
Hi sch,
R.M.,
Kundzewicz,
Z.W.,
Le -
enmaie ,
D.,
S ou e ,
R.J.,
2008.
S a iona i y
is
dead:
Whi he
wa e
managemen ?
Science
319,
573–574.
Mohammadi,
S.A.,
P asanna,
B.M.,
2003.
Analysis
o
gene ic
di e si y
in
c op
plan s
–
salien
s a is ical
ools
and
conside a ions.
C op
Sci.
43,
1235–1248.
Mulde ,
C.P.H.,
Bazeley-Whi e,
E.,
Dimi akopoulos,
P.G.,
Hec o ,
A.,
Sche e -Lo -
enzen,
M.,
Schmid,
B.,
2004.
Species
e enness
and
p oduc i i y
in
expe imen al
plan
communi ies.
Oikos
107,
50–63.
Naeem,
S.,
W igh ,
J.P.,
2003.
Disen angling
biodi e si y
e ec s
on
ecosys em
unc ioning:
de i ing
solu ions
o
a
seemingly
insu moun able
p oblem.
Ecol.
Le .
6,
567–579.
Nys o
¨m,
M.,
2006.
Redundancy
and
esponse
di e si y
o
unc ional
g oups:
implica ions
o
he
esilience
o
co al
ee s.
AMBIO
35,
30–35.
Page,
S.E.,
2010.
Di e si y
and
Complexi y.
P ince on
Uni e si y
P ess,
P ince on,
pp.
296.
Pe chey,
O.L.,
Gas on,
K.J.,
2006.
Func ional
di e si y:
back
o
basics
and
looking
o wa d.
Ecol.
Le .
9,
741–758.
Ro
¨ e ,
R.P.,
Ho
¨hn,
J.G.,
T nka,
M.,
F onzek,
S.,
Ca e ,
T.R.,
Kahiluo o,
H.,
2013.
Modelling
shi s
in
ag oclima e
and
c op
cul i a
esponse
unde
clima e
change.
Ecol.
E ol.
3,
4197–4214.
Shannon,
C.E.,
Wea e ,
W.,
1949.
The
Ma hema ical
Theo y
o
Communica ion.
Uni e si y
o
Illinois,
Illinois,
USA.
Sche e ,
M.,
Ca pen e ,
S.R.,
Foley,
J.A.,
Folke,
C.,
Walke ,
B.,
2001.
Ca as ophic
shi s
in
ecosys ems.
Na u e
413,
591–596.
Schindle ,
D.W.,
1990.
Expe imen al
pe u ba ions
o
whole
lakes
as
es s
o
hypo heses
conce ning
ecosys em
s uc u e
and
unc ion.
Oikos
57,
25–41.
T nka,
M.,
Olesen,
J.E.,
Ke sebaum,
K.C.,
Skjel a,
A.O.,
Ei zinge ,
J.,
Seguin,
P.,
Pel onen-Sainio,
P.,
Ro
¨ e ,
R.P.,
Iglesias,
A.,
O landini,
S.,
Dub o sky,
M.,
Hla-
inka,
P.,
Balek,
J.,
Ecke s en,
H.,
Cloppe ,
E.,
Calanca,
P.,
Gobin,
A.,
Vuce ic,
V.,
Nejedlik,
P.,
Kuma ,
S.,
Lalic,
B.,
Mes e,
A.,
Rossi,
F.,
Kozy a,
J.,
Alexand o ,
V.,
Seme ado a,
D.,
Alud,
Z.,
2011.
Ag oclima ic
condi ions
in
Eu ope
unde
clima e
change.
Global
Change
Biol.
17,
2298–2318.
Walke ,
B.H.,
Kinzig,
A.,
Lang idge,
J.,
1999.
Plan
a ibu e
di e si y,
esilience,
and
ecosys em
unc ion:
he
na u e
and
significance
o
dominan
and
mino
species.
Ecosys ems
2,
95–113.
Wa d
J .,
J.H.,
1963.
Hie a chical
g ouping
o
op imize
an
objec i e
unc ion.
J.
Am.
S a .
Assoc.
48,
236–244.
Win ee,
R.,
K emen,
C.,
2009.
A e
ecosys em
se ices
s abilized
by
di e ences
among
species?
A
es
using
c op
pollina ion.
P oc.
Roy.
Soc.
B
276,
229–237.
Yeo,
D.,
T uxillo,
C.,
2005.
Applied
Clus e ing
Techniques
Cou se
No es.
SAS
Ins i u e
Inc.,
Ca y,
NC
27513,
USA.
Zhu,
Y.,
Chen,
H.,
Fan,
J.,
Wang,
Y.,
Li,
Y.,
Chen,
J.,
Fan,
J.,
Yang,
S.,
Hu,
L.,
Leungk,
H.,
Mewk,
T.W.,
Tengk,
P.S.,
Wangk,
Z.,
Mund ,
C.C.,
2000.
Gene ic
di e si y
and
disease
con ol
in
ice.
Na u e
406,
718–722.
H.
Kahiluo o
e
al.
/
Global
En i onmen al
Change
25
(2014)
186–193
193