A mosphe ic and Clima e Sciences, 2013, 3, 165-176
h p://dx.doi.o g/10.4236/acs.2013.32018 Published Online Ap il 2013 (h p://www.sci p.o g/jou nal/acs)
Obse ed Changes in Long-Te m Clima ic Condi ions
and Inne -Regional Di e ences in U ban
Regions o he Bal ic Sea Coas
Michael Rich e 1*, Sonja Deppisch1, Hans Von S o ch2
1U ban Planning and Regional De elopmen , Ha en Ci y Uni e si y, Hambu g, Ge many
2Ins i u e o Coas al Cesea ch, Helmhol z-Zen um, Gees hach , Ge many
Email: *[email p o ec ed]
Recei ed No embe 14, 2012; e ised Janua y 1, 2013; accep ed Janua y 26, 2013
Copy igh © 2013 Michael Rich e e al. This is an open access a icle dis ibu ed unde he C ea i e Commons A ibu ion License,
which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
ABSTRACT
This pape p esen s esea ch ou comes om an in es iga ion in o clima e change and u ban impac s on clima e de el-
opmen in u ban egions o he Bal ic Sea coas . The ci ies conside ed we e Ros ock and S ockholm, and hei su -
ounding egions. The objec i es we e: 1) o de e mine whe he signi ican changes in empe a u e and p ecipi a ion
ha e occu ed and, i so, o wha ex en ; and 2) o es ablish whe he he e is a no iceable u ban hea island e ec in
S ockholm and he medium-sized ci y o Ros ock. Clima ic ends we e de ec ed by linea eg ession and he Mann-
Kendall es . Di e en p ecipi a ion ends we e de ec ed o e he whole pe iod o obse a ion. A e age annual em-
pe a u es inc eased signi ican ly in bo h case s udies, pa icula ly om he 1970s wi h highes ends in win e and low-
es in au umn (Ros ock) and summe (S ockholm). Al hough changes in empe a u e ex emes we e de ec ed o bo h
egions, no o e all long- e m end o p ecipi a ion ex emes was obse ed. The a e age empe a u e in he ci y o
Ros ock (S ockholm) was app oxima ely 0.3˚C o 0.6˚C (1.2˚C) highe han in he su ounding u al a eas had seasonal
a ia ions, wi h maxima in he wa m season. The main ou comes we e ha signi ican changes in clima ic condi ions,
pa icula ly empe a u e pa e ns, ha e been occu ing in he case s udy egions since he 1980s, and ha he e is a con-
side able u ban hea island e ec in bo h S ockholm and Ros ock. This could encou age u ban planne s o conside spe-
ci ic clima ic condi ions and small-scale clima ic in luences also in ela i ely small coas al u ban conglome a es in mid
la i udes which can ollow om land use changes.
Keywo ds: U ban Hea Island; Clima e Change; U ban Clima e; Bal ic Sea; Ex eme E en s
1. In oduc ion
1.1. Clima e Change in he Bal ic Sea Region
The global clima e is changing. The wa ming end o
he en i e globe was 0.04˚C pe decade om 1850 o
2005. Wa ming has accele a ed since he 1980s o a em-
pe a u e inc ease o app oxima ely 0.17˚C pe decade [1].
The wa ming end in he Bal ic Sea Region, a 0.10/
0.07˚C (no h/sou h) pe decade, exceeds he global end.
This may be because he globe consis s mainly o he -
mally ine ocean, leading o he mo e apid wa ming o
land masses [2]. In his con ex , he Bal ic Sea Region
e e s o he ca chmen a ea o he Bal ic Sea, loca ed in
no h-eas e n Eu ope (see Figu e 1). In he pe iod o
1871 o 2004, he mos ob ious seasonal ends occu ed
in sp ing, wi h inc eases o 0.15˚C and 0.11˚C pe decade
in he no h and sou h, espec i ely. The leas wa ming
occu ed in summe : 0.06˚C/0.03˚C pe decade [2]. Tem-
pe a u e inc eases o abou 1˚C in he 20 h cen u y ha e
been epo ed in he d ainage basin o he Bal ic Sea [2].
Globally, minimum ai empe a u es ose o a g ea e
ex en han maximum ai empe a u es, meaning ha he
diu nal empe a u e ange dec eased [2]. Re e ence [3]
analysed da a conce ning he he mal g owing season in
he Bal ic egion, and ound ha i had ex ended by abou
one week since 1951. This phenomenon ollows a g ow-
ing eas -wes g adien be ween Denma k and Finland,
and highe in a-annual a iabili y in he wes [3]. Global
ends in p ecipi a ion beha ed di e en ly, depending on
la i ude. In no he n a eas abo e 30˚N, ain all inc eased
in he 20 h cen u y, while i dec eased in opical a eas
be ween 10˚N and 10˚S [1]. Signi ican inc eases in
*Co esponding au ho .
C
opy igh © 2013 SciRes. ACS
M. RICHTER ET AL.
166
Figu e 1. Map o he Bal ic Sea egion in No he n Eu ope
wi h i s d ainage basin ou line and he loca ion o he ci ies
o Ros ock and S ockholm.
ain all could be ound, o example, in he eas e n a eas
o No h and Sou h Ame ica, No he n Eu ope and Cen-
al and No h Asia. By con as , he Sahel egions, he
Medi e anean, Sou h Asia and sou he n A ica became
d ie [1]. Acco ding o da a by [4], mo e p ecipi a ion
was measu ed in he wa e shed o he Bal ic Sea in he
pe iod o 1976 o 2000 han om 1951 o 1975. In he
Bal ic egion, ain all inc eased mos isibly in Sweden
and on he eas coas in he win e and sp ing seasons [2].
In summe he e was inc eased ain all in he no h,
whils less p ecipi a ion ell in he sou he n pa o he
egion. O e all, i ually no long- e m end is isible
since ain all a ies bo h spa ially and empo ally [2].
1.2. U ban Hea Island and Clima e Change
I is a well-known ac ha a mosphe ic condi ions in
ci ies di e om hose in he su ounding coun yside [5].
The u ban hea island (UHI) e ec desc ibes he phe-
nomenon o di e ing empe a u e and o he clima ic
ac o s be ween u ban a eas and hei immedia e su -
ounding u al a eas. In many ci ies, his e ec is mo e
like a “hea a chipel” han an island [6], depending on he
u ban s uc u e. In addi ion, he e a e changes in ain all,
cloud co e , wind condi ions and ai pollu ion. These
di e ences a e p ima ily gene a ed in u ban a eas due o
la ge-scale building s uc u es, sealing and emissions
om anspo and indus y. Fo a mo e de ailed summa y
o he easons o special u ban clima ic condi ions see o
example [7]. The a e age annual ai empe a u e in la ge
ci ies is 1˚C o 2˚C highe han in he immedia e su -
ounding a ea. In excep ional cases, di e ences o mo e
han 10˚C can be de e mined in mega ci ies a ce ain
imes [8]. Due o he u ban clima e, in e -annual em-
pe a u e a iabili y ends o dec ease [9]. S able s a i i-
ca ion and low cloud co e p omo e he o ma ion o an
UHI (see [10]); high wind speeds educe i s in ensi y. Di -
e en dependencies o he UHI in ensi y on wind speed
we e ound by a ious au ho s [8,10-12] o example, ha
he UHI in ensi y depends on he in e se squa e oo o
wind speed [8,12]. Re e ence [8] also disco e ed a loga-
i hmic ela ion be ween he popula ion o a ci y and he
maximum UHI in ensi y. Following Oke’s ela ion, a me-
dium-sized ci y such as Ros ock, wi h app oxima ely
200,000 inhabi an s, should ha e a maximum UHI in en-
si y o abou 6˚C o 7˚C. The u ban clima e e ec will
impac on an inc easing numbe o people wo ldwide in
he 21s cen u y because o e hal o he wo ld’s popula-
ion was ci y-dwelle s in 2010, and his endency is
g owing [13]. Regional clima e models ha e e ealed ha
mo e se e e consequences can be expec ed conce ning
u u e clima e change in me opoli an a eas han in hei
su ounding a eas [14]. I is assumed ha he numbe o
summe days (Tmax ≥ 25˚C) and ho days (Tmax ≥ 30˚C)
will inc ease in many egions [14], and hence he a eas
whe e people su e om he mal s ess [15]. He e, hea
wa es play an impo an ole and a e e y likely o be-
come mo e common and se e e in Eu ope [16]. I socie-
ies a e no p epa ed o such changes, mo bidi y and he
mo ali y a e will inc ease due o he consequences o a
highe hea load [15], an example o which occu ed in
F ance in he summe o 2003 [17]. In Ge many, 3500
people died as a consequence o he 2003 hea wa e [18].
Ac oss Eu ope, he e we e abou 22,000 o 45,000 hea -
ela ed dea hs wi hin wo weeks o Augus [19]. These
consequences o clima e change may, howe e , be e-
duced by in oducing app op ia e adap a ion measu es o
ci ies. On he o he hand, cold s ess and mo ali y is likely
o dec ease especially in mid-la i udes [6] and in ci ies
due o he u ban clima e e ec s [20].
1.3. Aim o he Pape
To e alua e pas and cu en changes in clima ic condi-
ions in he case s udy egions, his pape p esen s he
esul s o analyses o measu ed me eo ological pa ame-
e s om di e en ime scales and loca ions. Annual,
seasonal, mon hly and daily se ies we e analysed o he
case s udy egions o in es iga e long- e m ends o
empe a u e and p ecipi a ion. The e ec s o UHI we e
in es iga ed by compa ing he measu ed da a o in-
ne -ci y wea he s a ions wi h nea by subu ban/ u al s a-
ions wi h a ime esolu ion o hou s and days.
The main esea ch ques ions we e:
How has he clima e ( empe a u e, p ecipi a ion) chang-
ed in he case s udy egions and a e he e egional di -
e ences in he case s udy egions, o example, be ween
Copy igh © 2013 SciRes. ACS
M. RICHTER ET AL.
Copy igh © 2013 SciRes. ACS
167
coas al and inland a eas?
Is he e a no iceable UHI e ec in he case s udy e-
gions, and unde which speci ic wea he condi ions do
he la ges empe a u e di e ences occu ?
2. Analysis o he Da a-S udy Cases Ros ock
and S ockholm
2.1. S udy A eas
Ros ock, a ci y on he Ge man Bal ic Sea coas (see Fig-
u es 1 and 2), is he la ges ci y in he Fede al S a e o
Mecklenbu g-Vo pomme n (Ge many). The ci y is he
comme cial cen e o Mecklenbu g-Vo pomme n, and
had app oxima ely 202,000 inhabi an s in 2010 (1990:
250,000). One ea u e o he ci y is he mou h o he i e
Wa now, called B ei ling. Wi h an a ea o abou 1 km2,
he B ei ling is loca ed in he middle o he ci y, wi h he
seapo . The u ban cen e is loca ed a he sou he n end
o he Wa now del a. The a eas wi h highes deg ee o
sealing a e o he wes o he Wa now, whe e mos in-
habi an s li e. Ros ock is ela i ely la ; i s highes ele a-
ion (app oxima ely 50 m abo e sea le el) is in he sou h-
eas . A e da a om Ge man Me eo ological Se ice
(DWD), he a e age annual empe a u e is 8.4˚C; he
a e age annual p ecipi a ion is 590 mm (1961-1990, s a-
ion Ros ock-Wa nemünde).
The ci y o S ockholm is he capi al o Sweden, lo-
ca ed be ween he Bal ic Sea in he wes and Lake
Malä en in he eas (see Figu es 1 and 2). Wi h an a ea
o 187 km2, he ci y is sligh ly la ge han Ros ock, bu
i s popula ion densi y is much highe (4.552 pe km2).
S ockholm has abou 850,000 inhabi an s. A e da a
om Swedish Me eo ological and Hyd ological Agency
(SMHI), he a e age annual empe a u e is 6.6˚C and he
a e age annual p ecipi a ion 540 mm (1961-1990, s a ion
S ockholm-Obse a o ielunden).
Figu e 2. Loca ion o measu emen si es o he case s udy egions Ros ock (54˚05'20''N, 12˚08'24''E) and S ockholm
(59˚19'30''N, 18˚3'0''E). S a ion names: Za—Za en in, Bo—Bol enhagen, Schw—Schwe in, Ki/P—Ki chdo /Poehl, Go—
Goldbe g, Gü—Gülzow, Ro-Wa—Ros ock-Wa nemünde, Ro-Ho—Ros ock-Holbeinpla z, Ro—Ros ock, Ro-S —Ros ock-
S u ho , G Lü—G oß Lüsewi z, La—Laage, Ba—Ba h, Te—Te e ow, Wa—Wa en, Tu-Ai—Tullinge Ai po , S -B —
ockholm-B omma, S -Ob—S ockholm-Obse a o ielunden, S -Ho—S enska-Höga na. S
M. RICHTER ET AL.
168
2.2. Measu emen Si es, Da a Collec ion and
P epa a ion
Homogenised mon hly da a o empe a u e and p ecipi a-
ion om DWD we e used o analyse he long- e m cli-
ma ic ends in he egion o Ros ock. All o he chosen
s a ions we e loca ed in Mecklenbu g-Vo pomme n,
wi hin 100 km o Ros ock (see Figu e 2). T ends o
ex eme alues we e examined using daily da a conce n-
ing p ecipi a ion and maximum and minimum empe a-
u es om a ious s a ions. This da a was also supplied by
DWD. In addi ion, da a om he ai moni o ing ne wo k
o Mecklenbu g-Vo pomme n was supplied by he S a e
O ice o En i onmen , Na u e Conse a ion and Geol-
ogy o examine he u ban e ec s on empe a u e, in pa -
icula . The s a ions we e close o he ci y cen e (Ros ock-
Holbeinpla z) loca ed nea a busy oad, o a u al si es
nea Ros ock (Ros ock-S u ho , Gülzow) su ounded by
ag icul u al a eas o g assland. The s a ion Ros ock-Wa -
nemünde is loca ed di ec ly behind he dunes o he Bal ic
Sea Coas . The UHI in ensi y was calcula ed om he
di e ence be ween he empe a u es eco ded a he wea-
he s a ion nea he ci y cen e (Ros ock-Holbeinpla z)
and hose a he u al s a ions (Ros ock-S u ho , Ros ock-
Wa nemünde, Gülzow). Da a o he S ockholm egion
(see Figu e 2) was p o ided by he Swedish Me eo o-
logical and Hyd ological Agency (SMHI). This da a
comp ised he long- e m daily se ies o S ockholm-Ob-
se a o ielunden 1756-2009 [21,22], om which e ec s
such as he UHI e ec (see [23]) we e emo ed. Tempe a-
u e ex emes we e calcula ed using he 3-hou a e age
da a om S ockholm-B omma and S enska-Höga na (1961-
2009) and by calcula ing he daily maximum and mini-
mum empe a u e. U ban e ec s on empe a u e we e
compu ed using he wea he s a ion a S ockholm-B om-
ma ai po as he u ban s a ion and Tullinge Ai po as he
u al s a ion. S ockholm-B omma was used as u ban s a-
ion ins ead o S ockholm-Obse a o ielunden because o
ee a ailabili y o da a om 3-hou ly measu emen s. The
ime se ies o e alua ing clima ic changes we e di ided
in o di e en no mal pe iods o 30 yea s’ leng h. Due o
he expec ed di e en esul s o bo h clima ic changes
and he UHI e ec a se e al imes o he yea , he ime
se ies we e di ided u he in o seasons by mon hs (DJF/
MAM/JJA/SON).
2.3. S a is ical Me hods
To check he da a supplied by he ai moni o ing ne wo k
o Mecklenbu g-Vo pomme n, ou lie s we e iden i ied
and emo ed by checking he da a se ies g aphically and
looking o implausible di e ences be ween measu ed
empe a u e alues. This da a was s a is ically e iewed
and compa ed wi h da a om DWD by calcula ing de-
sc ip i e s a is ics, co ela ion coe icien s and c oss-
co ela ions on a daily and hou ly basis be ween he ime
se ies. Va ious s a is ical me hods we e applied o de e -
mine and es he signi icance o long- e m ends. T ends
conce ning annual o seasonal empe a u e and p ecipi a-
ion we e calcula ed using linea eg ession wi h leas -
squa e i ing o ob ain a linea end pe ime in e al. The
signi icance o hese ends was es ed using he Mann-
Kendall es [24]. Due o he possible se ial co ela ion o
he ime se ies, he Mann-Kendall es may become libe al,
alsely ejec ing he null hypo hesis [25]. Fo his eason,
all ime se ies we e il e ed by a “p ewhi ening” app oach
[25] o elimina e he p oblems caused by se ial co ela ion.
These s a is ical analyses we e conduc ed using he
ANCLIM so wa e by Pe Š ěpánek [26] and SPSS S a-
is ics. Gliding linea ends o annual mean empe a u es
we e es ima ed by linea eg ession using mo ing 31-yea
windows. The ends in ex emes we e calcula ed using
he RClimDex Package o R. A e a simple quali y
con ol o he da a ( eplacemen o missing & un eason-
able alues, iden i ica ion o ou lie s), he p og amme
compu es all 27 co e indices ecommended by he CCl/
CLIVAR Expe Team o Clima e Change De ec ion
Moni o ing and Indices (ETCCDMI), as well as o he
empe a u e and p ecipi a ion indices wi h use -de ined
h esholds [27].
3. Resul s
This sec ion is di ided in o wo main pa s. In he i s pa ,
a desc ip ion is gi en o he long- e m clima ic changes in
empe a u e, p ecipi a ion and indices o ex emes in he
case s udy egions. In he second pa , he u ban impac s
on empe a u e and he dependence o he UHI on wind
speed and cloud co e a e explo ed.
3.1. Long-Te m Changes
Tempe a u e changes in he Ros ock egion we e exam-
ined wi h mon hly da a om he s a ions a Schwe in,
Ros ock-Wa nemünde (Figu e 3) and Te e ow. Rep e-
sen a i e o he Ros ock egion, Figu es 3-5 show e-
sul s om he Ros ock-Wa nemünde s a ion. Fo he
whole pe iods unde s udy, he e a e posi i e ends o-
cusing on an annual base o abou 0.25˚C/decade (Figu e
4(a)) (signi ican (p < 0.05) o Ros ock-Wa nemünde,
Schwe in). The la ges inc eases in a e age annual em-
pe a u e occu ed in he pe iod om 1976 o 2005, wi h
linea ends o 0.5˚C o 0.62˚C/decade. When ocusing
on seasonal ends, i becomes appa en ha au umn
(SON) wa med leas conside ably a all s a ions. The
s onges inc eases in seasonal empe a u e occu ed in
win e be ween 1976 and 2005, a Wa nemünde wi h a
end o 0.79˚C/decade (Figu e 4(a)). Tempe a u es in
sp ing (MAM) and summe (JJA) inc eased by nea ly
0.7˚C/decade a Wa nemünde. The e seems o be a s onge
Copy igh © 2013 SciRes. ACS
M. RICHTER ET AL. 169
Figu e 3. Annual a e age empe a u e (1947-2009) a Ro-
s ock-Wa nemünde wi h linea end om 1951 o 2005
and S ockholm-Obse a o ielunden wi h linea end om
1947 o 2009 and plo ed di e ences be ween hese annual
empe a u es.
(a) (b)
Figu e 4. Linea ends [˚C pe yea ] o obse ed annual
empe a u es a Ros ock-Wa nemünde wea he s a ion (a)
and S ockholm-Obse a o ielunden (b). T ends a e o di -
e en seasons o o he whole yea and, due o di e en
measu emen pe iods and da a a ailabili y, o di e en
pe iods.
Figu e 5. Gliding linea ends o he annual mean em-
pe a u e, es ima ed by 31-yea windows, o wea he s a ions
Ros ock-Wa nemünde and S ockholm-Obse a o ielunden
o he ime be ween 1951 and 2009.
inc ease in annual a e age empe a u es since he la e
1980s o all se ies (see Figu e 3, o ins ance). Fo
S ockholm, he long- e m se ies om 1756 o 2009 [21,
22] was used o examine ends in a e age annual em-
pe a u es. These esul s (Figu es 3 and 4(b)) also indi-
ca e an inc ease in a e age annual empe a u e in he la e
1980s. Du ing he pe iod om 1756 o 2009, sligh in-
c eases (0.03˚C o 0.07˚C/decade) occu ed h oughou
he yea and o all seasons wi h he excep ion o summe ,
whe e a dec ease o 0.04˚C/decade was ound. All ends
we e signi ican (p < 0.05) o his pe iod. No signi ican
ends we e de ec ed o he pe iod om he 1950s
(1950-1979). These di e ences anged om −0.17˚C/
decade in au umn o 0.2˚C/decade o he en i e yea . Fo
he pe iod a e 1980, ends ange om 0.06˚C/decade in
summe o 1.05˚C/decade in win e (Figu e 4(b)). Fo he
en i e yea , he end is signi ican wi h 0.67˚C/decade (p
< 0.05). Gliding linea annual mean empe a u e ends,
es ima ed by linea i in mo ing 31-yea windows (Fig-
u e 5), showed empo al a ia ions in he a e o change.
The longe ime se ies o S ockholm showed a ia ions
anging om −0.4 o +0.4˚C/decade un il he 1980s.
F om his poin in ime (1985 o Ros ock and 1989 o
S ockholm), he end alues exceeded his ange. A e
1972/1973, he a e was posi i e and accele a ed o e he
en i e pe iod, wi h a ew in e up ions. The maximum
linea 31-yea end was achie ed in bo h Ros ock and
S ockholm, a 0.75˚C/decade, in 1993 (pe iod 1978-
2003). Gliding linea ends we e simila o e he en i e
pe iod om 1951 o 2009. A e 1972, he wa ming end
a Ros ock-Wa nemünde was almos always highe .
The ime se ies o p ecipi a ion we e il e ed by a
“p ewhi ening” app oach [25] o elimina e he p oblems
caused by se ial co ela ion and linea ends we e calcu-
la ed wi h 10-yea unning means and co esponding
signi icance was de ec ed wi h he Mann-Kendall-Tes .
Fo any pe iod, he e was an o e all posi i e end o
all seasons and o he whole yea o all s a ions, as was
he case o empe a u e. The linea 31-yea gliding mean
p ecipi a ion end a Ros ock wea he s a ion (Ro in
Figu e 2) was nega i e un il 1975 (Figu e 6), a e
which he end was posi i e up o 2009. The e was a
maximum inc ease in 1986 (1971-2001) a 6.9 mm/a, and
dec easing bu posi i e ends he ea e .
Table 1 shows linea ends o obse ed 10-yea un-
ning means o annual and seasonal p ecipi a ion om
1951 o 2005. No signi ican ends we e de ec ed by he
Mann-Kendall Tes o he pe iod om 1951 o 2005 (p
≤ 0.05). Fo Schwe in, Ros ock and Ki chdo , p ecipi a-
ion inc eased by 0.32 mm/a o 0.49 mm/a. The la ges
inc eases in p ecipi a ion occu ed in win e and au umn;
by con as , dec eases o small inc eases we e mainly
eco ded in sp ing and summe . The annual linea p e-
cipi a ion end o Wa en was nega i e o he whole
Copy igh © 2013 SciRes. ACS
M. RICHTER ET AL.
170
(a)
(b)
Figu e 6. Annual p ecipi a ion a Ros ock wea he s a ion
wi h linea end (a) and 31-yea gliding linea end (b) o
1951 o 2005.
Table 1. Linea ends [mm pe yea ] o obse ed 10-yea
unning means o annual p ecipi a ion om 1951 o 2005.
T ends a e o di e en seasons and o he whole yea .
S a ion Yea DJF MAM JJA SON
Schwe in 0.48 0.52 0.135 −0.67 0.57
Ros ock 0.49 0.41 0.07 −0.09 0.63
Ki chdo 0.32 0.15 −0.003 0.29 0.32
Wa en −1.5 −0.16 −0.39 −0.86 −0.51
yea and o all seasons.
Clima e indices o he wea he s a ions we e calcu-
la ed om daily alues o maximum empe a u e, mini-
mum empe a u e and p ecipi a ion. Table 2 shows al-
ues wi h signi ican (p < 0.05) inc easing o dec easing
ends. I can be seen ha he leng h o he he mal
g owing season (GSL) inc eased signi ican ly o all s a-
ions. The numbe o os days pe yea (FD0) dec eased,
and he numbe o summe days (SU25) inc eased sig-
ni ican ly a mos loca ions.
O he indices show only a ew signi ican dec eases o
inc eases a a ious si es. Schwe in is he only s a ion
wi h an inc ease in ex eme p ecipi a ion indices, R20
mm (numbe o e y hea y p ecipi a ion days, annual
coun o days when p > 20 mm) and R95 pp ( e y we
days, annual o al p ecipi a ion when p > 95 h pe cen ile).
Indices o he S ockholm egion we e compu ed using
da a om S ockholm-B omma and S enska-Höga na (Ta-
ble 3).
He e, he numbe o ice days pe yea (ID0) and he
numbe o os days pe yea (FD0) dec eased signi i-
can ly, and he mon hly minimum alue o daily mini-
mum empe a u e (TNn) and he mon hly maximum
alue o daily minimum empe a u e (TNx) inc eased
signi ican ly a bo h si es om 1961 o 2009. Cold ex-
emes (FD0, ID0) dec eased by abou 0.5 o 1 day pe
yea . Addi ionally, he numbe o summe days (SU25),
he leng h o he g owing season (GSL) and he mon hly
maximum and minimum alue o daily maximum em-
pe a u e (TXx, TXn) inc eased a S enska Höga na.
3.2. U ban Impac s
Fo he Ros ock egion, da a wi h a ime s ep o one hou
we e a ailable o Ros ock-Wa nemünde (HRO_Wa ne-
muende), Ros ock-Holbeinpla z (HRO_Holbeinpla z),
Ros ock-S u ho (HRO_S u ho ) and Guelzow (see Fig-
u e 2). Table 4 shows desc ip i e s a is ics o he hou ly
empe a u e measu emen s. The s a ion nea he ci y cen-
e o Ros ock (HRO_Holbeinpla z) had he highes mean
empe a u e (10.2˚C) du ing he pe iod om 2001 o
2009.
Ros ock-Holbeinpla z and Ros ock-Wa nemünde had
conside ably highe absolu e minimum empe a u es
(−13.3˚C and −13.9˚C). The mean empe a u e di e ence
be ween he s a ion in he inne ci y (HRO-Holbeinpla z)
and u al s a ions nea he ci y, o he mean UHI e ec , is
be ween 0.3˚C and 0.6˚C. The same alues a e p oduced
when ocusing on daily a e age empe a u es (Table 5).
He e, he mean empe a u e di e ence be ween HRO-
Holbeinpla z and G oss Lüsewi z wea he s a ion (G _
Luesewi z, u al s a ion, dis ance abou 15 km) was e en
1.4˚C. The a e age diu nal cycle (Figu e 7) shows ha a
nigh , HRO_Wa nemünde had he highes empe a u es,
which we e abou 1˚C highe han a HRO_S u ho and
Guelzow a 5:00 h s. HRO_Holbeinpla z was wa me han
hese s a ions, bu abou 0.5˚C colde han a HRO_
Wa nemuende a nigh . A 9:00 h s he mean empe a u es
we e he same; empe a u es a HRO_Holbeinpla z in-
c eased mo e apidly han a he o he loca ions un il
14:00/15:00 h s, when he a e age di e ences be ween
he nea es s a ions—HRO_S u ho and HRO_Wa ne-
münde—we e abou 1˚C o 2˚C. F om la e a e noon
onwa ds, empe a u es dec eased un il hey again eached
Copy igh © 2013 SciRes. ACS
M. RICHTER ET AL.
Copy igh © 2013 SciRes. ACS
171
Table 2. Signi ican (p ≤ 0.05) changes in clima ic co e indces ( ollowing [27]) in days/yea o ˚C/yea o a ious si es in and
a ound Ros ock.
Clima e indices Wa ne-Münde Te e ow Schwe-Rin G oss Lüse-Wi z Goldbe g Bol enhagen Ba h
DTR 0.018 −0.009 0.013
FD0 −0.385 −0.146 −0.568 −0.432 −0.49
ID0 −0.241 −0.2
GSL 0.91 0.668 0.335 0.901 0.901 0.758 0.79
R20 mm 0.009
R95 pp 0.35
SU25 0.169 0.329 0.268 0.209 0.16
TNn 0.05 0.083 0.064
TNx 0.028 0.011 0.002 0.022
TXx 0.058 0.051
TR20 0.015
Table 3. Signi ican (p ≤ 0.05) changes in clima ic co e indi-
ces ( ollowing [27]) in days/yea o ˚C/yea o a ious si es
in and a ound S ockholm.
Clima e indices S ockholm-B omma S enska Höga na
SU25 0.056
ID0 −0.485 −0.667
FD0 −0.531 −0.985
GSL 0.499
TXx 0.078
TXn 0.084
TNx 0.028 0.048
TNn 0.092 0.096
Table 5. Desc ip i e s a is ics o s a ions including em-
pe a u e ange, absolu e minimum and maximum empe a-
u e, mean empe a u e and s anda d de ia ion o daily
da a measu ed 2001-2009 in ˚C.
S a ion Range Min Max Mean S d de
HRO_Holbeinpla z 38.8 –11.0 27.8 10.2 7.1
HRO_S u ho 37.7 –11.4 26.3 9.6 6.8
HRO_Wa nemünde 36.2 –10.7 25.5 9.9 6.8
Laage 38.7 –12.2 26.5 9.3 7.1
G _Luesewi z 37.8 –12.4 25.4 8.8 6.9
Guelzow 38.7 –11.6 27.1 9.9 7.1
Table 4. Desc ip i e s a is ics o s a ions including em-
pe a u e ange, absolu e minimum and maximum empe a-
u e, mean empe a u e and s anda d de ia ion o hou ly
da a measu ed 2001-2009 in ˚C.
S a ion Range Min Max Mean S d de
HRO_Holbeinpla z 49.3 –13.9 35.4 10.2 7.6
HRO_S u ho 54.2 –19.2 35.0 9.6 7.4
Guelzow 54.4 –18.4 36.0 9.9 7.7
HRO_Wa nemuende 46.3 –13.3 33.0 9.9 7.0 Figu e 7. A e age diu nal empe a u e cycles o di e en
s a ions 2001-2009.
M. RICHTER ET AL.
172
a poin when hey we e nea ly he same, a 20:00/21:00
h s.
This pa e n emained he same h oughou he yea .
Howe e , empe a u e di e ences be ween HRO_Hol-
beinpla z and HRO_Wa nemünde we e lowe a nigh
(<0.6˚C) and highe in he ea ly a e noon (up o 2.5˚C),
pa icula ly in sp ing (MAM) and summe (JJA, Figu e
8(b)). The di e ences in empe a u e o e he day in
win e (DJF, Figu e 8(a)) and au umn (SON) we e less
conside able. When compa ing he empe a u e di e -
ences a HRO_Holbeinpla z and HRO_S u ho , he high-
es empe a u e di e ence was always in he ea ly a e -
noon. In summe he e was ano he UHI maximum a
midnigh and wo minima a 7:00 h s and 19:00 h s.
Minima o empe a u e di e ences be ween HRO_Hol-
beinpla z and Guelzow occu ed in he la e a e noon in
summe and win e ; he summe maximum was in he
ea ly mo ning (5:00/6:00 h s) and he win e maximum
was a midday. Figu e 9 shows ha he mon hly mean
UHI was highes in May, wi h a di e ence o o e 1˚C o
(a)
(b)
Figu e 8. A e age diu nal cycle o UHI in ensi y o win e
mon hs ((a), DJF) and summe mon hs ((b), JJA) o 2001-
2009. UHI in ensi ies we e compu ed om each di e ence
be ween Ros ock-Holbeinpla z (Ho) and Ros ock-S u ho ,
Ros ock-Wa nemünde and Gülzow wea he s a ions.
Figu e 9. A e age mon hly UHI in ensi y o 2001 o 2009.
UHI in ensi ies we e compu ed om each di e ence o
mon hly a e ages be ween Ros ock-Holbeinpla z (Ho) and
Ros ock-S u ho , Ros ock-Wa nemünde and Gülzow wea-
he s a ions.
HRO_Wa nemünde and HRO_S u ho , and lowes in
win e . When HRO_Holbeinpla z was compa ed wi h he
coas al s a ion HRO_Wa nemünde, a e age empe a u es
nea he coas we e highe in au umn and win e . The e is
no such clea seasonal cycle o he di e ence be ween
HRO_Holbeinpla z and Gülzow bu wo maxima in
summe and ea ly win e and wo minima in au umn and
a he end o win e . The maximum UHI in ensi y o e he
obse ed pe iod was 8.5˚C o HRO_S u ho and 8.3˚C
o Guelzow o hou ly measu emen s and 4.4˚C and
5.7˚C, espec i ely, o daily mean empe a u es. Da a
om wo s a ions was compa ed o he S ockholm egion.
S ockholm-B omma is a wea he s a ion a S ockholm-
B omma Ai po o he no hwes o he ci y cen e;
Tullinge Ai po is a mo e u al s a ion sou h o S ock-
holm. Th ee-hou ly empe a u e da a was a ailable o he
pe iod om 1996 o 2009 o bo h si es, which we e hen
compa ed. In his pe iod, S ockholm-B omma had an
a e age empe a u e o 7.4˚C and Tullinge Ai po 6.2˚C.
This would be a mean UHI o 1.2˚C, compa able, o
example, o Hambu g [12] and London [28]. The maxi-
mum UHI in ensi y o 3-hou ly measu emen s and daily
means (in b acke s) is 12.9˚C (8.3˚C), he maximum U -
ban cool island (UCI) in ensi y is −5.4˚C (−2.4˚C), e-
spec i ely. As in Ros ock, he diu nal cycle o he UHI
changed h oughou he yea . As can be seen in Figu e 10,
he win e UHI was only weak, a e aging a be ween
0.4˚C a midday and 0.8˚C be o e midnigh . The e we e
signi ican diu nal a ia ions o he UHI in ensi y du ing
he o he seasons. The a e age UHI was a i s maximum a
a ound midnigh , and was highes in summe , a abou
2.3˚C. The a e age UHI in ensi ies dec eased om 3:00
h s o 6:00 h s; he minimum le el was a a ound 9:00 h s,
a e which UHI in ensi ies inc eased again un il a ound
midnigh .
Copy igh © 2013 SciRes. ACS
M. RICHTER ET AL. 173
Figu e 10. A e age diu nal cycle o UHI in ensi y o he
whole yea , win e mon hs (DJF), sp ing mon hs (MAM),
summe mon hs (JJA) and au umn mon hs (SON) o 1996
o 2009. UHI in ensi ies we e compu ed om he di e ence
be ween a e ages a S ockholm-B omma and Tullinge-Ai -
po .
The dependence o he UHI in ensi y on di e en me-
eo ological condi ions has been epo ed in a ious
s udies [10,12,29]. He e i was examined whe he he e
was a signi ican ela ionship be ween he daily mean UHI
in ensi y (daily mean ΔT be ween Ros ock-Holbeinpla z
and Ros ock-S u ho ) in Ros ock and he condi ions o
wind speed and cloud co e . Ros ock-Wa nemünde wea-
he (synop ic s a ion) s a ion was used as he e e ence
s a ion o he me eo ological condi ions due o he eli-
abili y o i s da a. Because o he e en ual dis up i e in-
luences o .e. land-sea b eeze ci cula ion on wind speed
measu emen , he analysis was pe o med wi h daily wind
speed da a o o he synop ic s a ions nea Ros ock
(Schwe in, Goldbe g), oo. A linea eg ession analysis
wi h he leas squa e me hod was ca ied ou o all da a
and seasons wi h Equa ion (1):
UHI a x b (1)
whe e x is he daily mean o wind speed o cloud co e .
Fo wind speed, he pa ame e s a e gi en in Table 6. All
eg essions a e signi ican (p < 0.05). Pa ame e s o he
o he s a ions we e e y simila in o de o magni ude
and seasonali y and a e no shown he e. The coe icien
o de e mina ion 2 anged om 0.11 in win e o 0.20 in
au umn. The alue b de e mines he heo e ical daily a -
e age UHI in ensi y when he a e age daily wind speed is
0. This anges om 0.42˚C in win e o 1.53˚C in sp ing.
The s onge de e mined a iance o UHI in ensi y can
be explained by cloud co e (Table 7, all signi ican o
p < 0.05). The 2 alues a e mos ly highe han hose o
wind speed excep o win e and au umn. Fo cloud
co e , he s onges dependence o UHI in ensi y is du -
ing sp ing and summe ; he leas dependence is in win e .
The heo e ical UHI in ensi y when he e was no cloud
co e is 0.3˚C in win e and 1.81˚C in summe .
Table 6. Pa ame e s o linea eg ession be ween daily a -
e age UHI in ensi y ( empe a u e di e ence be ween Ro-
s ock-Holbeinpla z and Ros ock-S u ho ) and wind speed a
Ros ock-Wa nemünde.
a b 2
Yea –0.11 1.08 0.16
DJF –0.05 0.42 0.11
MAM –0.15 1.53 0.19
JJA –0.12 1.40 0.16
SON –0.01 0.74 0.20
Table 7. Pa ame e s o linea eg ession be ween daily a -
e age UHI in ensi y ( empe a u e di e ence be ween Ro-
s ock-Holbeinpla z and Ros ock-S u ho ) and cloud co e
(in %) a Ros ock-Wa nemünde.
a b 2
Yea –0.02 1.36 0.22
DJF 0.00 0.30 0.02
MAM –0.02 1.63 0.25
JJA –0.02 1.81 0.27
SON –0.01 0.83 0.14
4. Discussion
T end analyses o long- e m p ecipi a ion and empe a-
u e on an annual and seasonal basis e ealed signi ican
changes in clima ic condi ions in bo h case s udy egions,
Ros ock and S ockholm. All o he examined ime se ies
o empe a u e measu emen s showed sus ained accele -
a ed wa ming since he 1980s compa ed o ea lie pe iods.
This is in line wi h inc eased wa ming on a global scale
[30] and in he Bal ic Sea ca chmen [2] since he 1980s.
Bo h s udy a eas wa med mos in win e mon hs (Ros ock:
0.7˚C o 0.79˚C/decade, S ockholm: 1.05˚C/decade). This
is in line wi h he BACC-Repo [2], which s a ed ha he
wa ming ends a e highe in no he n la i udes. Di e -
ences exis in he leas wa ming pe iods. The Ros ock
egion wa med leas in au umn (−0.1˚C o 0.26˚C/decade);
o he S ockholm egion, he lowes wa ming end
(0.06˚C/decade) was in summe . Ou analysis o he p e-
cipi a ion ime se ies e ealed no clea inc easing o de-
c easing end in he Ros ock egion, bu a s ong a i-
abili y in annual p ecipi a ion quan i ies. The long- e m
ends we e pa ly signi ican , bu only in he o de o
magni ude o 1‰ o annual p ecipi a ion quan i ies. The
sho e 31-yea gliding linea end a Ros ock wea he
s a ion was up o he o de o magni ude o 1% o he
annual p ecipi a ion quan i y. Th ee o he ou s a ions
ha e nega i e ends in summe ; as expec ed, he wo
mos con inen al s a ions, Schwe in and Wa en, had he
Copy igh © 2013 SciRes. ACS
