Climate change is implicating a two-fold impact on air temperature increase in the ripening period under the conditions of the Luxembourgish grapegrowing region

Abstract

Aim: Grape (Vitis vinifera L.) phenology is mainly temperature-driven. Consequently, the shift in thermal conditions due to climate change is supposed to have a distinct influence on grape phenology, grape maturity and wine typicity. This study aims to investigate (i) the future phenological development, as well as (ii) the consequences on the temperature conditions in specific phenophases under the conditions of the Luxembourg grapegrowing region.
Methods and Results: A budburst model and a phenological model were combined with an ensemble of ten regional climate change projections for Luxembourg. Analyses comparing four 30-year time spans (reference period: 1971-2000; present: 2001-2030; near future: 2031-2060; far future: 2061-2090) demonstrated that each of the 27 phenological stages according to BBCH code is projected to be reached significantly earlier than in the reference period. According to these projections, the length of phenophases at the early stages is not affected, whereas the ripening period length is significantly shortened. The air temperature increase in the ripening period (far future compared to reference period: + 4.6 °C to + 5.3 °C) is projected to be markedly higher than in the annual averages (+ 2.6 °C).
Conclusions: Since (i) air temperatures are generally projected to increase in the future and (ii) the ripening period will take place earlier (usually in the warmer parts of the season), climate change is implicating a two-fold impact on air temperature increase in the ripening period.
Significance and impact of the study: This two-fold impact potentially threatens the wine typicity of the traditional grapegrowing regions and therefore calls for specific adaptation strategies.

Full text

Climate change is implicating a two-fold impact
on air temperature increase in the ripening period under the conditions
of the Luxembourgish grapegrowing region
Daniel Molitor*and Jürgen Junk
Luxembourg Institute of Science and Technology (LIST), “Environmental Research and Innovation (ERIN)”
Department, 41, rue du Brill, L-4422 Belvaux, Luxembourg
Corresponding author : daniel.molitor@list.lu
Aim: Grape (Vitis vinifera L.) phenology is mainly temperature-driven. Consequently, the shift in thermal
conditions due to climate change is supposed to have a distinct influence on grape phenology, grape maturity and
wine typicity. This study aims to investigate (i) the future phenological development, as well as (ii) the
consequences on the temperature conditions in specific phenophases under the conditions of the Luxembourg
grapegrowing region.
Methods and Results: A budburst model and a phenological model were combined with an ensemble of ten
regional climate change projections for Luxembourg. Analyses comparing four 30-year time spans (reference
period: 1971-2000; present: 2001-2030; near future: 2031-2060; far future: 2061-2090) demonstrated that each of
the 27 phenological stages according to BBCH code is projected to be reached significantly earlier than in the
reference period. According to these projections, the length of phenophases at the early stages is not affected,
whereas the ripening period length is significantly shortened. The air temperature increase in the ripening period (far
future compared to reference period: + 4.6 °C to + 5.3 °C) is projected to be markedly higher than in the annual
averages (+ 2.6 °C).
Conclusions: Since (i) air temperatures are generally projected to increase in the future and (ii) the ripening period
will take place earlier (usually in the warmer parts of the season), climate change is implicating a two-fold impact on
air temperature increase in the ripening period.
Significance and impact of the study: This two-fold impact potentially threatens the wine typicity of the traditional
grapegrowing regions and therefore calls for specific adaptation strategies.
climate change, cumulative degree days, multi-model ensemble, phenology, ripening temperature
AB S T R A C T
K E Y W O R D S
Received: 9 October 2018
y
Accepted: 12 April 2019
y
Published: 8 July 2019
DOI: 10.20870/oeno-one.2019.53.3.2329
VINE AND WINE
OPEN ACCESS JOURNAL
409
OENO One 2019, 3, 409-422 © 2019 International Viticulture and Enology Society - IVES
Additional tables can be downloaded from https://oeno-one.eu/article/view/2329

INTRODUCTION
The grapevine (Vitis vinifera L.) is a perennial
plan t with an an nua l cycle that is hi ghly
dependent on environmental conditions (Parisi et
al., 2014) . Am ong the m, air tempe rat ure
represents a central factor for the cultivation of
grapevines. In cases where the water, nutrient
and radiation requirements of the plants are
fulfilled (Nendel, 2010, Webb et al., 2007),
thermal conditions control “with only minor
other influences” (Gladstones, 2011) grapevine
phe nolo gy and final grape ma tur ity.
Consequently, changes in thermal conditions are
impacting grape phenology, grape maturity, wine
typicity and, as a last consequence, the economic
sustainability of many traditional grapegrowing
regions such as Luxembourg, where the wine
industry traditionally represents an economically
important sector.
On the east- to south-faced vineyards along the
Moselle River grapevines have been cultivated
since the Roman times. The quality and th e
qua ntit y of a nnua l win e pro d uct ion i n
Luxembourg have been documented in a wine
chronicle since the beginning of the 9th century
A.C. (Molitor et al., 2016b). Nowadays (2017),
the tot al area of the Luxem bou rgi s h
grapegrowing regions covers 1303 ha between
Sc hen gen and R osp ort ove r approxim ate ly
42 km along the Moselle River (Anonymous,
2018). The position of the viticultural fields in
the Luxem bourgish grap egrowing regio n is
depicted in Figure 1. In 2017, most cultivated
cultivars were Müller-Thurgau (syn. Rivaner)
(23.3% of the total acreage), Pinot gris (15.2%),
Auxerrois (14.8%), Pinot blanc (12.6%),
Riesling (12.5%), Pinot noir (9.7%) and Elbling
(6.1%) (Anonymous, 2018). Total wine
pr oduct ion of t he r egion on ave rage of the
vi nta ges 2 008 to 20 17 reached 109 0 92 hl
(Anonymous, 2018).
Worldwide, thermal conditions have significantly
changed during recent decades at both global and
regional scale. According to the Intergovern-
mental Panel on Climate Change (IPCC), human
influence, primarily the burning of fossil fuels,
has been the dominant cause of global warming
for several decades (IPCC, 2013). Based on
regional climate change projections taken from
the ENSEMBLES and the CORDEX projects, an
air te mper atur e incre ase of up to 4 °C –
dep endi ng o n th e emi ssio n sc enar io – is
projected for Luxembourg by the end of this
century (Goergen et al., 2013; Junk et al., 2016).
Also, changes in precipitation patterns towards
drier summers and wetter winters are projected
(Goergen et al., 2013).
Sin ce g rape ph enol ogic al dev elop ment is
predomi nantly air temperature-d riven (e.g.,
Duchene and Schneider, 2005; Gladstones, 2011;
Keller, 2015; Moncur et al., 1989), the projected
temperature increase due to climate change will
hav e sig nifi can t imp acts on vi ticu ltur e,
particularly close to the climatic frontiers of
vi tic ult ure w her e the dep end enc e of gr ape
phenology and maturity on climatic conditions is
mos t p rono unc e d ( Bra z dil et a l., 2008).
Ph enolo gy represen ts a maj or f actor in the
distribution of the viticultural areas (Garcia de
Cortazar-Atauri et al., 2017). With ongoing
climatic change, northern regions are expected to
become more suitab le (in terms of climatic
co ndi tions ) for rip ening g rapes ( Jones a nd
Sc hul tz, 2 016 ). In co ntr ast , reg ion s where
temperatures are already close to optimum for
be st wi ne q ualit y might bec ome too hot to
produce high quality wines with balanced fruit in
the future (Jones and Schultz, 2016). To assess
the viti cult ural oppo rtun itie s, ris ks an d
cha llen ges rel ate d to cl imat e chang e and
fu rther more to sup port th e de velop ment of
adequate adaptation strategies, numerical models
simulating the effects of temperature conditions
on plant development are helpful tools.
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
410
FIGURE 1. The Luxembourgish grapegrowing
region. Vineyards are depicted in red, main rivers
in blue and borders between countries in black.
The weather station used for present
investigations is located in Remich.
4

Recently, Molitor et al. (2014b) developed a
high-resolution phenology model covering all
27 B B CH (Bio logi sch e Bu ndes anst alt ,
Bundessor-tenamt und Chemische Industrie)
plant phenological growth stages defined by
Lorenz et al. (1995) from budburst to grape
harvest (i.e., BBCH stage 89 describes “grapes
ripe for harvest”). The incorporation of (i) an
upper threshold temperature, above which a
further increase of the te mp er ature will not
accelerate plant development, and (ii) a heat
threshold, above which a further increase of the
temperature will slow down plant development,
in this model have been demonstrated to improve
the p r eci sion of the mo del compa red to
commonly used un-capped cumulative degree
day-based phenology models (e.g., Amerine and
Winkler, 1944; Duchene et al., 2010; Hoppmann,
2010; Nendel, 2010; Oliveira, 1998; Parker et
al., 2011; Schultz, 1992; Zapata et al., 2015).
Und er inc reas ed air t empe ratu re, th e
improvement in precision gained through the
inc orpo rat ion of add itio nal t hres hold s is
expected to be even more pronounced (Molitor
et al., 2014b).
Sin ce (i) air temp erat ure s are exp ect ed to
increase in the future due to climate change and
(ii) the ripening period will likely take place
earlier, usually in the warmer parts of the season
due to faster phenological development driven
by higher temperatures, a two-fold impact on air
temperature increase in the ripening period could
be expected (Duchene et al., 2010), while little is
known about the influence of climate change on
the temperature conditions in other phenophases.
In addi tion , a temp erat ure inc reas e in the
ripening period is expected to alter the typicity of
the wine (Jackson and Lombard, 1993).
Consequently, the aim of the present study was
to in ves tiga te the f utu re phe nolo gica l
development of the cultivars Müller-Thurgau,
Rie slin g and Pin ot noir, as wel l as the
consequences on the air temperature conditions
in spe cif ic phe nop hase s, esp eci ally i n the
ripening period, based on (i) the budburst model
of Molitor et al. (2014a), (ii) the high-resolution
phenological model of Molitor et al. (2014b) and
(iii) a multi-model ensemble of ten regional
climate change projections under the conditions
of the Luxembourgish grapegrowing region.
MATERIALS AND METHODS
1. Observation data: daily mean air
temperatures Remich 1970-2016
Air temperature data were recorded from 1970
thr ough 201 6 b y a wea the r s tati on of the
Luxembourgish national agricultural adminis-
tr ation ASTA ( Administrati on d es s ervic es
techniques de l’agriculture) located in the centre
of the Luxembourgish grapegrowing region in
Remich/Luxembourg (49.54° N, 6.35° E; 207 m
a.s.l.) (Figure 1). Unventilated air temperatures
were measured at 2 m above the ground. Daily
mea n air tem pera tur e s wer e c alc ulat ed as
av era ges o f daily m ini mum and m axi mum
temperatures.
2. Modelled data: daily mean air
temperatures 1970-2090
Time s eries o f daily m ean air t emp era ture
between 1970 and 2090 were extracted from the
online archives of the EU ENSEMBLES project
(http://ensembles-eu.metoffice.com/). In order to
assess the uncertainties related to climate change
projecti on s, a multi-model ense mb le of ten
regional climate change projections was used,
bas ed on th e A1B e miss ion sc enar io
(Supplementary Table 1).
The A1B e miss ion s cena rio d escr ibes
anthropogenic emissions of a future world with
rapid economic growth until the middle of this
century and a balanced use of fossil and non-
fossil energy resources (Nakicenovic and Swart,
2000). It is widely used in impact assessments
for Central Europe (Junk et al., 2015a; Junk et
al., 2015b; Junk et al., 2016; Lokys et al., 2015;
Molitor et al., 2014a).
The selected ensemble covers the overall range
of the available regional climate models (RCMs)
in terms of air temperature change signals and
accounts for the most widely used European
RCMs (van Pelt et al., 2012). Time series of
daily data for Remich were extracted from each
RCM. Instead of using the information from just
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 411
TABLE 1. Optimized threshold temperatures for degree day accumulation as well as average coefficients
of variance in the three cultivars Müller-Thurgau, Riesling and Pinot noir.
Cultivar
Lower threshold
(°C)
Upper threshold
(°C)
Heat threshold
(°C)
Ave ra ge
coefficient
of variance
Reference
Müller-Thurgau 5 20 22 0.1473 Molitor et al. (2014b)
Riesling 7 18 24 0.1465 Molitor et al. (2016)
Pinot noir 3 20 24 0.1572

one single grid cell of the model results, a spatial
mean of 3 × 3 grid cells around that central point
(Remich) was used (spatial resolution of 25 km
× 25 km per grid cell) (Goergen et al., 2013;
Junk e t al ., 20 12; Matza rakis et al. , 2013)
(Supplementary Figure 1). Regional climate
models show syst ema tic d iff ere nce s whe n
compared to direct point measurements. In our
study the impact models for budburst and the
BBCH stages are both based on absolute values
and therefore it is necessary to apply a bias
correction. We used long-term measurements
from the Remich site for the bias correction. The
applied method of quantile mapping is described
in detail in Molitor et al. (2014a) and Junk et al.
(2015b). Correction factors were calculated for
the period 1971-2000 and then applied to the
period of investigation from 1970 to 2090.
3. Phenological models
3.1. Budburst model
The budburst date (expressed as day of the year
– DOY) was cal cul ated fo r each en semble
me mber and each yea r ba sed on the mod el
developed by Molitor et al. (2014b) for the
Müller-Thurgau cultivar. This model represents a
parameterization of the DORMPHOT model
(Caffarra et al., 2011) simulating budburst for
photoperiod-sensitive plant species. It considers
(i) the dormancy induction process occurring in
late summer-autumn; (ii) the action of chilling
temperatures for dormancy release; and (iii) the
promoting effect of a long photoperiod on bud
development during dormancy release and bud
development (Caffarra and Eccel, 2011).
2. High-resolution cumulative degree day-based
models to simulate the phenological
development
The dates of reaching all 27 phenological stages
(according to the BBCH scheme (Lorenz et al.,
19 95)) bet ween bu dburs t an d ha rvest were
calculated for the three Vitis vinifera cultivars
Mül ler-Thu rga u, Rie slin g an d Pinot noir,
according to the high-resolution cumulative
degree day-based phenological model (Molitor
et al., 2014b). In contrast to linear cumulative
degree day approaches, this model takes into
co nsiderat ion that th e fo rcing e ff ect of ai r
temperature is limited at higher temperatures by
incorporating (i) an upper threshold, above
which a further increase of the temperature will
not accelerate plant development, and (ii) a heat
threshold, above which a further increase of the
temperature will slow down plant development
(Molitor et al., 2014b). Investigations of Molitor
et al. (2014b) demonst rated that o ptimized
tem per atur e thr esh olds f or veg eta tive and
generative development are almost identical with
those determined for the whole phenological
cycle. Hence, a single model covering the whole
phenological development was used according to
Mol itor et al. (2 0 14a ). Mos t adeq uat e
tem pera tur e t hres hold s to simu lat e t he
phenological development were selected based
on minimum average (average of all stages)
co effic ients o f va riation (CV; the standa rd
deviation divided by the mean) of the cumulative
degree days (Molitor et al., 2014).
The optimized thresholds (leading to minimum
coefficient of variation) for Müller-Thurgau and
Riesling were taken from Molitor et al. (2014b)
and Molitor et al. (2016), respectively (Table 1).
For Pinot noir, a parameterization took place
following the approach of Molitor et al. (2014b)
bas ed on 2 6 lon g-t e rm phen olog ical and
meteorolo gic al observa tio n dat a set s fro m
El tvill e (G ermany), Kin del (Germ any) and
Remich (Luxembourg).
Threshold triplets (cardinal temperatures) with
best predictive fit were determined based on the
coefficients of variance on the average of all
stages as described before (Molitor et al., 2014b;
Mol ito r et al. , 201 6). For P ino t noir, bes t
adaptation on the 26 long-term phenological data
sets (lowest average coefficient of variation
(0.1 57 2)) was achieved using the th reshold
triplet 3°C, 20°C and 24°C.
An overview of the optimized thresholds for
deg ree day ac c umu lati on in the diff ere nt
cultivars is given in Table 1.
Average (representing the average value of the
26 data sets) cumulative degree days reaching
specific BBCH stages for all three cultivars are
given in Supplementary Table 2.
The budburst date according to the budburst
model represents the starting date of the high-
resolution cumulative degree day-based model.
An aly ses of multi- ann ual obs erv ation d ata
recorded in Remich/Luxembourg (Supplemen-
tary Table 3) demonstrate that the DOYs of
budburst (BBCH 09) of Riesling and Pinot noir
do not differ significantly from the DOYs in
Müller-Thurgau (according to non-parametric
paired-sample t-test; p≤ 0.05). Hence, calculated
dates of budburst for Müller-Thurgau according
to Molitor et al. (2014b) were used for all three
cultivars.
A ‘phenophase’ is defined as the time span
between reaching a specific BBCH stage and
reaching the subsequent stage (e.g., time span
BBCH 09 to 11 = phenophase 09).
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
412

4. Determination of dates of reaching
phenological stages
Bas ed o n (i) the b udb u rst mo del , (ii) the
phenological model and (iii) the ensemble of ten
regional climate change projections, the DOYs
for reaching each of the 27 phenological stages
between budburst and harvest were calculated
for each year, each of the ten ensemble members
and each of the three cultivars. The average dates
(30 years 10 projections - n = 300) of all stages
in the subseq uent 30 -year time spans wer e
computed for all cultivars:
- the reference period (“past”) from 1971 to
2000,
- the “present” from 2001 to 2030,
- the “near future” from 2031 to 2060 and
- the “far future” from 2061 to 2090.
5. Determination of air temperatures in
different phases
Every year, for each of the ten regional climate
cha nge pro jec tion s and eac h of the t h ree
cultivars of investigation, the subsequent average
air temperatures were calculated:
- annual and monthly temperatures,
- pre-bloom temperature (budbreak (BBCH 09)
to beginning of flowering (BBCH 61)),
- bloom temperature (beginning of flowering
(BBCH 61) to end of flowering (BBCH 69)),
- post-bloom temperature (end of flowering
(BBCH 69) to veraison (BBCH 81)) and
- ripening temperature (veraison (BBCH 81) to
berries ripe for harvest (BBCH 89)).
Air temperatures in the four phases defined
above were calculated as mean air temperatures
in the period between the DOY after reaching
the starting stage (BBCH 09, BBCH 61, BBCH
69, BBCH 81, respectively) and the DOY of
reaching the terminal stage (BBCH 61, BBCH
69, BBCH 81, BBCH 89, respectively) for each
projection, each year and each cultivar. Average
air temperatures (30 years 10 projections - n =
300) were calculated in the four 30-year time
spans defined above and for the phases for each
cultivar.
Additionally, the average number of years, in
which the BB CH stage 89 “berries ripe f or
ha rve st” was n ot re ach ed un til 3 1/10, w as
computed for each 30-year time span. In the
event that BBCH 89 (“berries ripe for harvest”)
was not reached by 31/10 (number of cases see
Supplementary Table 4), the terminal date for
ripening temperature calculation was fixed at
DOY 304 (31/10) to avoid the impact of low
temperatures after the vegetation period (such as
in November or December) on the calculated
average ripening temperature.
Moreover, the average daily air temperatures
were computed for all four time spans. Their
annual course was plotted (i) relative to 01/01 as
well as (ii ) relative to t he date of budburst
(BBCH 09).
6. Statistical analyses
Data sets consisting of 300 data per time span
(30 year s * 10 r egio nal cl imat e c hang e
projections) under present (2001-2030), near
future (2031-2060) and far future (2061-2090)
conditions were generally tested for significant
differences compared to the reference period
(past; 1971-2000) by non-parametric Mann-
Whi tney U- t est (p ≤ 0. 001) , us ing SPSS
Statistics 19 (IBM, Chicago, IL, USA). For the
phenological data sets, analyses were conducted
separately for each cultivar.
RESULTS
1. Annual temperature evolution
Figure 2 shows the observed annual mean air
temperatures for the weather station at Remich
as well as the multi-model mean of the te n
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 413
FIGURE 2. Observed annual average
temperatures in Remich (red line) and projected
(A1B emission scenario; ten ensemble-based
regional climate change projections; multi-model
mean) annual average temperatures (blue line) in
the period 1970 to 2100. Ensemble spread (+/- 1
standard deviation) is indicated in grey.

ens embl e mem b ers ( bias c orre cted ). A
sig nifi can t inc rea se in the ann ual air
temperatures compared to the reference period is
projected (Table 2).
2. Projected average phenological dates
The DOYs of all 27 phenological stages are
modelled to occur significantly earlier in all
cultivars in the present, near future and far future
compared to the reference period (Figures 3
to 5). This shift in time increases continuously
from the present to the far future. The temporal
difference compared to the reference period
al rea dy exi sts a t bu dbu rst (BBCH 09 ) and
remains relatively constant until the beginning of
the ripening period (Supplementary Tables 5
to 7). Consequently, in the time span 2001-2030,
no sig nificant cha nges in the len gth of th e
different phenophases between BBCH 09 and
BBCH 77 were projected in comparison to the
reference period. In contrast, the phenophase 85
(period between BBCH 85 and BBCH 89) is
modelled to get significantly shorter in all three
cultivars. The decrease in the length of this
phenophase compared to the reference period is -
3.9 (Müller-Thurgau), -3.5 (Riesling) and -5.7
(Pinot noir) days in the present, -7.7 (Müller-
Thurgau), -7.7 (Riesling) and -10.2 (Pinot noir)
day s i n the ne ar f utur e and -9.3 (Müll er-
Thurgau), -9.7 (Riesling) and -13.0 (Pinot noir)
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
414
TABLE 2. Average (ten ensemble-based regional climate change projections) annual air temperatures,
monthly air temperatures, as well as pre-bloom (BBCH 09-61), bloom (BBCH 61-69), post-bloom
(BBCH 69-81) and ripening (BBCH 81-89) air temperatures in the Müller-Thurgau, Riesling and Pinot
noir cultivars in the different 30-year time spans. Δ (°C) = temperature difference (in Kelvin) compared to
the reference period (past; 1971-2000).
Past (1971-2000)
T (°C) T (°C) ! (°C) T (°C) ! (°C) T (°C) ! (°C)
Year (01/01-31/12)10.310.8 0.5 11.8 1.5 12.9 2.6
January 2.3 2.9 0.6 4.3 2.0 5.4 3.1
February 3.0 3.7 0.7 4.9 1.9 5.8 2.8
March 5.9 6.6 0.7 7.4 1.5 8.4 2.5
April 9.6 10.2 0.6 11 .1 1.5 11.4 1.8
May 13.8 14.2 0.4 15.0 1.2 15.6 1.8
June 16.8 17.3 0.5 18.1 1.3 19.4 2.6
July 18.9 19.5 0.7 20.2 1.3 21.4 2.5
August 18.3 19.0 0.7 19.9 1.7 21.1 2.8
September 15.2 15.7 0.5 16.6 1.4 17.8 2.6
October 10.8 11.2 0.4 12.2 1.4 13.3 2.5
November 5.6 6.1 0.6 7.3 1.8 8.7 3.2
December 2.7 2.9 0.2 4.3 1.5 5.9 3.1
Müller-Thurgau
Pre-bloom (BBCH 09-61) 14.4 14.4 0.0 14.6 0.2 14.5 0.1
Bloom (BBCH 61-69) 17.6 17.8 0.1 18.2 0.5 18.3 0.6
Post-bloom (BBCH 69-81) 18.8 19.3 0.5 19.7 0.8 20.7 1.8
Ripening (BBCH 81-89) 16.3 17.5 1.2 19.2 2.9 20.9 4.6
Riesling
Pre-bloom (BBCH 09-61) 14.5 14.6 0.0 14.7 0.2 14.7 0.1
Bloom (BBCH 61-69) 17.7 17.9 0.2 18.4 0.7 18.4 0.7
Post-bloom (BBCH 69-81) 18.7 19.3 0.6 19.9 1.2 20.9 2.2
Ripening (BBCH 81-89) 15.2 16.6 1.3 18.4 3.2 20.3 5.1
Pinot noir
Pre-bloom (BBCH 09-61) 14.4 14.4 0.0 14.5 0.2 14.5 0.1
Bloom (BBCH 61-69) 17.5 17.6 0.1 18.0 0.5 18.2 0.6
Post-bloom (BBCH 69-81) 18.7 19.3 0.5 19.8 1.0 20.7 2.0
Ripening (BBCH 81-89) 15.0 16.5 1.5 18.4 3.3 20.3 5.3
Present (2001-2030)
Near future (2031-2060)
Far future (2061-2090)
Temperatures of the same phases that differed significantly according to the non-parametric Mann-Whitney U-test (p= 0.001) compared to the
reference period (1971-2000) are marked in bold.

day s in th e far futu re (Fig ures 3 to 5;
Supplementary Tables 5 to 7).
The average number of years per 30-year time
span, in which stage BBCH 89 was not reached
until 31/10, is projected to decrease from 1.3
(Müller-Thurgau), 4.7 (Riesling), and 3.8 (Pinot
noir) in the reference period to 0.1 (Müller-
Thurgau), 1.2 (Riesling), and 1.1 (Pinot noir) in
the present. In the near future and far future, no
such cases we re ob serv ed (Suppl eme nta ry
Table 4).
3. Temperature conditions in different
phenophases
Annual and monthly average air temperatures
(ex cept f or Jan uary, M a y, Octo ber and
December) a re modelled to be significan tl y
higher in the 2001-2030 time span than in the
reference period. In the near and the far future
annual as well as all monthly temperatures are
projected to be significantly higher than in the
ref eren ce peri od. The c ompu ted ann ual
temperature increase compared to the reference
period is 0.5 °C (present), 1.5 °C (near future)
and 2.6 °C (far future) (Table 2).
In all three cultivars, no significant differences
were projected in the pre-bloom temperatures
compared to the reference period. On the other
hand, the post-bloom and ripening temperatures
are modelled to increase significantly in the
present, near future and far future. The projected
increase (in comparison to the reference period)
is most pronounced in the ripening period. Here,
compared to the reference period, temperatures
are modelled to increase from 1.2 °C (Müller-
Thurgau), 1.3 °C (Riesling) and 1.5 °C (Pinot
noir) in the present to 2.9 °C (Müller-Thurgau),
3.2 °C (Riesling) and 3.3 °C (Pinot noir) in the
near future, to 4.6 °C (Müller-Thurgau), 5.1 °C
(Riesling) and 5.3 °C (Pinot noir) in the far
future (Table 2; Figure 6).
DISCUSSION
Present analyses revealed a constant increase of
the annu al ai r te mper ature in the
Luxembourgish grapegrowing region in the
future confirming previous studies (Junk et al.,
2015a, Junk et al., 2015b, Junk et al., 2016,
Lokys et al., 2015, Molitor et al., 2014b).
The sim ula tion of the fu tur e phe nol ogical
development took p lace based on the high-
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 415
FIGURE 3. Days of the year (DOY) reaching the
phenological stages 09 to 89 according to the
BBCH scale (Lorenz et al., 1995) in the Müller-
Thurgau cultivar in the four 30-year time spans.
The box plots indicate the medians and the 25%
and 75% percentiles, whiskers are limited to one
standard deviation. Box plots of the days of the
year of the same BBCH stage that differed
significantly according to the non-parametric
Mann-Whitney U-test (p= 0.001) compared to
the reference period (1971-2000) are marked in
red.
FIGURE 4. Days of the year (DOY) reaching the
phenological stages 09 to 89 according to the
BBCH scale (Lorenz et al., 1995) in the Riesling
cultivar in the four 30-year time spans. The box
plots are indicating the medians and the 25% and
75% percentiles, whiskers are limited to one
standard deviation. Box plots of the days of the
year of the same BBCH stage that differed
significantly according to the non-parametric
Mann-Whitney U-test (p= 0.001) compared to
the reference period (1971-2000) are marked in
red.

resolution phenological model as proposed by
Molitor et al. (2014b). Here, optimized threshold
temperature triplets for phenology simulation are
defined statistically per cultivar based on long-
observation data. Even though these optimized
thresholds are purely statistical, model validation
for Müll er- Thu rga u over a broa d ran ge of
locations in Europe showed a high accuracy
(Molitor et al., 2014). Using this cumulative
degree day approach, in all three cultivars of
investigation, all 27 phenological stages were
computed to be reached significantly earlier in
the future time spans than in the reference period
confirming studies in other viticultural regions in
recent years (e.g., Caffarra and Eccel, 2011;
Duchene and Schneider, 2005; Duchene et al.,
2010; Fraga et al., 2016; Garcia de Cortazar-
Atauri et al., 2017; Ramos, 2017; Sadras and
Moran, 2013; Trought et al., 2015). Differences
in t he e xte n t o f ear lier appe aran ce of
phenological stages as well as in the length of
phenophases observed by the different authors
are supposed to be caused by differences in (i)
the phenological models used (e.g., taking into
account the effect above optimum temperatures
or not), (ii) cultivars of investigation as well as
(iii) t he climatic conditions in t he studied
regions.
In the past, BBCH stage 89 (“berries ripe for
harvest”) was simulated not to be reached in all
years in all three cultivars. This was especially
the case for the (compared to Müller-Thurgau)
relatively late ripening cultivars Pinot noir and
Riesling. In the near as well as in the far future,
heat summatio n will – according to present
analyses – not be a limiting factor for full grape
maturity – not even in Riesling or Pinot noir.
Due to generally increasing air temperatures,
budburst (BBCH 09) is modelled to be reached
significantly earlier in the future than in the past
confirming previous analyses (e.g., Caffarra and
Eccel, 2011; Fila et al., 2012, Molitor et al.,
2014a). The consequences of the earlier budburst
on the future late frost risk are controversially
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
416
FIGURE 5. Days of the year (DOY) reaching the
phenological stages 09 to 89 according to the
BBCH scale (Lorenz et al., 1995) in the Pinot
noir cultivar in the four 30-year time spans. The
box plots are indicating the medians and the 25%
and 75% percentiles, whiskers are limited to one
standard deviation. Box plots of the days of the
year of the same BBCH stage that differed
significantly according to the non-parametric
Mann-Whitney U-test (p= 0.001) compared to
the reference period (1971-2000) are marked in
red.
FIGURE 6. Average (ten ensemble-based regional climate change projections) pre-bloom (BBCH 09-
61), bloom (BBCH 61-69), post-bloom (BBCH 69-81) and ripening (BBCH 81-89) air temperatures in
the cultivars Müller-Thurgau (left), Riesling (centre) and Pinot noir (right) in the different 30-year time
spans. *= temperatures of the same phases that differed significantly compared to the reference period
(1971-2000) according to non-parametric Mann-Whitney U-test (p= 0.001).

discussed in scientific literature (Kartschall et
al., 2015, Kotremba et al., 2014, Molitor et al.,
2014a, Mosedale et al., 2015, Sgubin et al.,
2018 ). While the an alyses of Molitor et a l.
(2014a) indicate that the late frost risk might
decrease in the future, other analyses revealed
inconsistent or even increasing late frost risks.
These contrary results might be explained by the
respective underlying budburst models used.
The total length of the period between BBCH 09
and BBCH 89 (season duration) is projected to
be shor tene d in the futu re in all cul tiva rs
confirming projected data of Webb et al. (2007)
for th e Au stra lia n gr apeg row i ng regi ons.
Int eres tin gly, th e len gth of the diff ere n t
phenophases prior to veraison is projected not to
change significantly in the future compared to
the reference period. That is, the shift of the
phenological development until veraison towards
the beginning of the year is mainly the result of
the earlier budburst. This is in accordance with
observations of Duchene and Schneider (2005)
in the Alsace region where the time span from
budburst to flowering was constant between
1965 and 2003. In contrast, the phenophase 85
representing the period between BBCH 85 and
BBCH 89 is projected to be up to 13 days shorter
in the far future in the case of Pinot noir. The
explanation for adverse effects on phenophase
lengths in different phases of development can
be attributed to the course of the daily average
air temperatures in the four time spans. Figure 7
demonstrates that this increase is relatively
constant in the course of the year (left) compared
to t he re fere nce p eri od. I n con tras t, when
pl otting th e daily av erage air tempera tures
relatively to the date of budburst, comparable
temperature conditions are projected in the
period around BBCH 09 in all four time spans
(Fi gure 7; ri ght) . As conse quen ce of the
relatively constant temperature conditions, the
length of the phenophases (mainly determined
by temperature conditions) is not significantly
affected in the early stages but shifted towards
the beginning of the year.
As a consequence, the computed pre-bloom air
tem pera tur es do not show any s igni fica nt
differences between the reference period and the
future time spans. This is the case because this
phase is shifted towards the earlier (usually
colder) part of the year. In consequence, mainly
tem pera tur e-de pend ent ste ps o f g rape
physiology between budburst and bloom (such
as inf loresce nce differe ntiatio n an d fl ower
initiation (Keller 2015; Molitor and Keller,
2016)) might not be systematically affected
(eve n though there is an increas e in spring
temperatures).
Whi le, ac cord ing to pr esen t resu lts, ai r
temperatures are already increasing significantly
in the post-bloom to veraison period, the length
of the phenophase stays relatively constant until
veraison. This apparent contradiction can be
explained by the fact that the air temperature
conditions in this period are situated in all time
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 417
FIGURE 7. Course of daily average temperatures in the four time spans (past: 1971-2000; present: 2001-
2030; near future: 2031-2060; far future: 2061-2090) (i) relative to 01/01 (day of the year – DOY; left) or
(ii) relative to the date of budburst (BBCH 09; right).

spans predominately in the optimum range for
further phenological development (e.g., in the
case of Riesling between 18 °C and 24 °C daily
mean air temperature). Here, a further slight
increase of daily average air temperatures does
no t influ ence the pace of t he ph enolo gical
dev elop men t, i f the hea t t h res hold is not
exceeded.
For practical viticulture under given climatic
conditions, the relatively stable length of the
phe noph ase s prio r to ver aiso n mean s tha t
consequences on the temporal distribution of the
pr e-v era iso n workload i n the vineyard are
expected to be relatively slight. That is, since
phenological development is not accelerated in
this period, spray intervals and time frames for
canopy management measures, for example,
might not be affected.
However, higher air temperatures in the time
frame between bloom and veraison might have
an impact on annual yield. Recent studies have
demonstrated negative correlations between
yield and temperatures, especially maximum
temperatures, over the first three weeks after
bloom for the Riesling cultivar (Molitor and
Keller, 2016). Hence, higher air temperatures in
this period might lead to a decrease in annual
yield in the future while in other periods the
effect of higher temperatures on the annual yield
was observed to be mainly positive (Molitor and
Keller, 2016). To clarify the overall effect of
climate change on yield formation, in a further
step, the present data on future temperature and
phenological conditions will be combined with
the temperature- and precipitation-driven yield
models for the Müller-Thurgau and Riesling
cultivars developed based on the multi-annual
yie ld re cord s for t he Lux emb o urg ish
grapegrowing region (Molitor and Keller, 2016).
The most distinct influences on the temperature
con dit ions as well as on t he le ngth o f the
phenophases are modelled for the ripening
pe rio d between verais on an d harvest. T he
increase in the ripening temperature is most
pronounced in late ripening cultivars such as
Riesling and Pinot noir. Here, the grapes are
ri pen ing at a lat er st age of the year where
tem pera tur e dif fer e nce s com pare d to t he
reference period temperatures are most distinct
(temperature decrease in the autumn), while in
case of Müller-Thurgau, the ripening period is
closer to the summer plateau of air temperatures.
The p r oje cted inc reas e in the ripe nin g
temperatures in the near future is 1.9 °C (Müller-
Thurgau; 2.9 °C), 2.1 °C (Riesling; 3.2 °C) and
2.2 °C times higher (Pinot noir; 3.3 °C) than the
temperature increase in the month of September
(1.5 °C). This phenomenon is the result of two
additional effects: (i) the general temperature
incr ea se and (ii) t he shift of t he phenolo gy
towards the earlier, generally warmer period of
the year.
In fact, this two-fold effect demonstrates that
changes in temperature conditions in calendar-
based (=anthropogenic) time frames (such as in
the “Cool Night Index ” ( Tonie tto an d
Carbonneau, 2004), taking into account
min imum tem per atur es in th e mo nth of
September (northern hemisphere)), might not
completely reveal the real temperature changes
in specific developmental stages.
More generally, this fact is indicating a general
limitation of calendar-based climatic indices
used in viticulture such as the Winkler index
(Amerine and Winkler 1944), the heliothermic
index of Huglin (1978) or the average growing
season temperature according to Jones (2007)
since they do not take into account the plant
response to climate (Caubel et al., 2015) and,
hence, might not per fectly re flect th e rea l
veg etat ion pe riod of sp ecif ic cul tiva rs
(Holzkämper et al., 2010).
Present results furthermore demonstrate that the
length of the ripening period (BBCH 81-89) is
projected to decrease in the future (e.g., Riesling,
past: 44.6 days; far future: 33.9 days). This is
confirming the results of the analyses of Tomasi
et al. (2011) in Northern Italy. Authors observed
a significant decrease of the length of the period
between veraison and harvest in the period 1964
to 200 9 ca used by inc reasi ng t emper ature s
(Tomasi et al., 2011). Due to the fact that the
harvest date might under practical conditions be
inf luen ced b y se v era l fac tors (Gar cia d e
Co rtazar-At auri et al., 201 0) the imp act of
temperature on the length of the ripening period
is controversially discussed in the literature. For
example, v an Le euw en and De strac Ir vin e
(2017) found opposite trends in the length of the
period between veraison and harvest. However,
based on present results we assume that the
temperature-sum approach proposed here might,
at least under the climatic conditions of the
Luxembourgish grapegrowing region, better
reveal the real average temperature conditions in
the simulated ripening period (since its length is
ass umed to be in flue nce d by tem per atur e
conditions) than approaches which calculate the
temperature conditions in this period as the
average temperatures between veraison and a
tem pora lly fixe d d ate such as ( i) 35 days
thereafter (Duchene et al., 2010) or (ii) 60 days
thereafter (Schultz and Hofmann, 2017).
Th e str ong a ir te mpe rat ure i ncr eas e in t he
ripening period as described above is presumably
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
418

linked to distinct changes in the wine typicity of
a specific region.
In fac t, hi ghe r te mpe ratur es ar e leadi ng to
altering fruit ripening rates (Martinez-Lüscher et
al., 2015), changes in flavour and aroma profiles
(Trought et al., 2015) as well as decoupled
an thocyanin and sugar syn theses in be rries
(Sadras and Mora n, 2013). Expected higher
su gar c onc ent rat ions ar e lea din g to h igh er
alcohol contents (Jackson and Lombard, 1993) in
the wines, and an acceleration of the degradation
of o rga nic acid s (Du chen e et al., 2010),
threatening both the freshness and lightness that
is esp ecial ly ex empla ry f or w hite wines in
(former) cool climate grapegrowing regions,
such as Luxembourg. Furthermore, the fruitiness
and aroma of grapes and wines is expected to be
neg ativ ely affect ed by high ri peni ng
temperatures (Duchene et al., 2010). To maintain
the wine ty pici ty o f the regio n, pot ent ial
ada ptat ion s tra tegi es to miti gate the
consequences of climatic change in general and
higher ripening temperatures in particular might
consist of measures leading to a temporal delay
of the maturation period. This might be achieved
by a shift towards cooler sites (e.g., with higher
elevations or lower exposition) or regions (e.g.,
at higher latitudes), cultivars or clones with a
later ripening characteristic, maturity-retarding
rootstocks, the application of antitranspirants
(Gatti et al., 2016) or specific crop cultural
measures including training systems (Molitor et
al., 2019), delayed winter pruning (Friend and
Trought, 2007) and adapted canopy management
(Parker et al., 2016, Stoll et al., 2013, Trought et
al., 2015).
CONCLUSIONS
Present analyses demonstrated that under the
cli mat ic con dit ions i n the L uxem bou rgi sh
grapegrowing region each of the 27 phenological
stages according to BBCH code are projected to
be reached significantly earlier in the future than
in the r efer enc e p eri od. Whil e sign ific ant
changes in the phenophase lengths are absent in
early stages , the r ipe nin g period le ngth is
significantly shortened in the future according to
these projections. Since (i) air temperatures are
generally projected to increase in the future and
(ii) the ripening period will take place earlier
(usually in the warmer parts of the season),
climate change is implicating a two-fold impact
on ripening period air temperature increase.
Consequently, the air temperature increase in the
ripening period (far future compared to reference
period: + 4.6 °C to + 5.3 °C) is projected to be
markedly higher than in the annual averages (+
2.6 °C). This significant increase of the ripening
period air temperatures potentially threatens the
wine typicity of the traditional grapegrowing
reg ions an d th ere fore cal ls for spe cifi c
adaptation strategies.
Ac knowled gements : The authors tha nk B.
Fuchs (Weinbauamt Eltville, Germany), O. Baus
(Hochschule Geisenheim University), S. Fischer,
R. Ma nnes and M. Schul tz ( Inst itu t Viti-
Vinicole, Remich, Luxembourg) for providing
par ts of the h ist oric al mete orol ogi cal and
phen ological d ata sets , F.K. Ronel lenfitsch
(L IST ) fo r GIS map s, M . Sulis ( LIST) f or
critical proof-reading, L. Auguin (LIST) for
language editing, B. Augenstein and R. Krause
(Geosens Ingenieurpartnerschaft, Schallstadt,
Germany) for running the phenological models
on the VitiMeteo platform, J. Niewind (LIST)
and P. Sinigoj (former CRP – Gabriel Lippmann)
for their support in data management, the Institut
Viti -Vinic ole for f inan cia l supp ort in the
framework of the research project “TerroirFuture
– Impact of climate change on viticulture in
Lu xem burg: r isk -as ses sme nt an d pot ent ial
adaptation strategies” as well as the European
Union in the framework of the “Clim4Vitis”
research project (Horizon 2020 research and
innovation programme; grant agreement No.
810176).
REFERENCES
Amerine M.A. and Winkler A.J., 1944. Composition
and quality of musts and wines of California grapes.
Hilgardia 15, 493-675. doi:10.3733/hilg.v15n06p493
Anonymous, 2018. Das Weinjahr 2017 und seine
Ernteergebnisse. Veröffentlichung der Abteilung
Weinbau des Weininstitutes in Remich.
Br azdil R ., Zahr adn ice k P ., Do bro vol ny P.,
Kotyza O. and Valasek H., 2008. Historical and
re ce nt viticulture as a sour ce of climato lo gi ca l
knowledge in the Czech Republic. Geografie-Sbornik
CG 113, 351-371.
Caffarra A., Donnelly A., Chuine I. and Jones M.B.,
2011. Modelling the timing of Betula pube scens
budburst. I. Temperature and photoperiod: a
conceptual model. Climate Research 46, 147-157.
doi:10.3354/cr00980
Ca ff arra A . and Eccel E., 20 11. Proj ecting the
impacts o f climate change on th e phenology of
grapevine in a mountain area. Australian Journal of
Grape and Wine Research 17, 52-61. doi:10.1111/
j.1755-0238.2010.00118.x
Caubel J., Garcia de Cortazar-Atauri I., Launay M.,
de Noblet-Ducoudré N., Huard F., Bertuzzi P. and
Graux A.-I., 2015. Broadening the scope for
ec oclimatic in dic ato rs to as ses s crop cl ima te
suitability according to ecophysiological, technical
an d qu ali ty cri ter ia. Ag ricultu ral an d Fore st
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 419

Meteorology 207, 94-106. doi:10.1016/j.agrformet.
2015.02.005
Duchene E., Huard F., Dumas V., Schneider C. and
Merdinoglu D., 2010. The challenge of adapting
gr apevine varieti es to cli ma te cha ng e. Cli ma te
Research 41, 193-204. doi:10.3354/cr00850
Duchene E. and Schneider C., 2005. Grapevine and
climatic changes: a glance at the situation in Alsace.
Agronomy for Sustainable Development 25, 93-99.
doi:10.1051/agro:2004057
Fila G., Di Lena B., Gardiman M., Storchi P., Tomasi
D., Silvestroni O. and Pitacco A., 2012. Calibration
and validation of grapevine budburst models using
growth-room experiments as data source. Agricul-
tu ral and F ore st Met eor olo gy 160, 69-79.
doi:10.1016/j.agrformet.2012.03.003
Fraga H., Santos J. A., Moutinho-Pereira J., Carlos
C. , Sil vestr e J., Eir as D ias J. E., M ota T. and
Malheiro A., 2016. Statistical modelling of grapevine
phenology in Portuguese wine regions: observed
trends and climate change projections. The Journal of
Agricultural Science 154, 795-811. doi:10.1017/
S0021859615000933
Friend A.P. and Trought M.C.T., 2007. Delayed
winter spur-pruning in New Zealand can alter yield
components of Merlot grapevines. Australian Journal
of Grape and Wi ne Res ear ch 13, 157-164.
doi:10.1111/j.1755-0238.2007.tb00246.x
Garcia de Cortazar-Atauri I., Daux V., Garnier E.,
Yiou P., Viov y N., Seguin B., Boursiqu ot J.M.,
Parker A.K., van Leeuwen C. and Chuine I., 2010.
Climate reconstructions from grape harvest dates:
methodology and uncertainties. Holocene 20, 599-
608. doi:10.1177/0959683609356585
Garcia de Cortazar-Atauri I., Duchene E., Destrac
Irvine A., Barbeau G., de Resseguier L., Lacombe T.,
Parker A.K., Saurin N. and van Leeuwen C., 2017.
Grapevine phe nol ogy in F ran ce: fro m pa st
observations to future evolutions in the context of
cl imate cha nge . Oen o One 51, 115-126.
doi:10.20870/oeno-one.2017.51.2.1622
Gatti M., Galbignani M., Garavani A., Bernizzoni F.,
To mbesi S., P all iot ti A . and P oni S ., 2016 .
Manipulation of ripening via antitranspirants in cv.
Barbera (Vitis vinifera L.). Australian Journal of
Grape and Wine Research 22, 245-255. doi:10.1111/
ajgw.12212
Gladsto nes J., 2011. Wine, Terroir and Climate
Change (Wakefield Press, Kent Town).
Goergen K., Beersma L., Hoffmann L. and Junk J.,
2013. ENSEMBLES-based assessment of regional
climate effects in Luxembourg and their impact on
vegetation. Cli mat ic C han ge 119, 761-773.
doi:10.1007/s10584-013-0756-x
Holzkämper A., Calanca P. and Fuhre r J., 2010.
Evaluating climate suitability for agriculture based on
agr ocl im ati c in dices . In: Swayne D., Yang W.,
Voinov A., Rizzoli A. and Filatova T. (eds.) 2010
International Congress on Environmental Modelling
and Software Modelling for Environment’s Sake.
Ottawa, C ana da (I nte rna tio nal En vir onm ent al
Modelling and Software Society).
Hoppmann D., 2010. Terroir. Wetter, Klima, Boden
(Verlag Eugen Ulmer, Stuttgart).
Huglin M.P., 1978. Nouveau mode d’évaluation des
possibilités héliothermiques d’un milieu viticole.
Compt es-rendus de l’Académie d’Agriculture de
France 64, 1117-1126.
IPCC, 2013. Climate Change 2013: The Physical
Science Basis. Contribution of Working Group I to
the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change (Cambridge University
Press, Cambridge). https://www.ipcc.ch/report/ar5/
wg1/
Jackson D.I. and Lombard P.B., 1993. Environmental
an d m ana gem ent p rac tices aff ect ing g rape
composition and wine quality - a review. American
Journal of Enology and Viticulture 44, 409-430.
Jones G.V., 2007. Climate change: observations,
projections, and general implications for viticulture
and wine production. In:Essi ck E., Gr iffin P. ,
Ke efer B., Mille r S. and Sto rchma nn K. (ed s.)
Working Paper No. 7. Walla Walla, WA: Whitman
College Economics Department.
Jones G.V. and Schultz H.R., 2016. Climate change
and emerging cool climate wine regions. Wine &
Viticulture Journal 31, 51-53.
Junk J., Eickermann M., Goergen K., Beyer M. and
Hoffm ann L., 201 2. Ensem ble-based analysis of
regional climate change effects on the cabbage stem
weevi l (Ceutorhynchus pallidactylus (Mrsh.)) in
winter oilseed rape (Brassica napus L.). The Journal
of Agricultural Science 150, 191-202. doi:10.1017/
S0021859611000529
Junk J. , Jo nas M . an d Eic ker man n M. , 201 5a.
Assessing meteorological key factors influencing crop
invasion by pollen beetle (Meligethes aeneus F.) -
pa st obs erv ati ons an d fut ure persp ect ive s.
Meteorologische Zeitschrift 25, 357-364. doi:10.1127/
metz/2015/0665
Junk J., Kouadio L., Delfosse P. and El Jarroudi M.,
2016. Effects of regional climate change on brown
rust disease in winter wheat. Climatic Change 135,
439-451. doi:10.1007/s10584-015-1587-8
Junk J., Ulber B., Vidal S. and Eickerma nn M.,
2015b. Assessing climate change impacts on the rape
stem weevil, Ceutorhynchus napi Gyll., based on
bias- and non-bias-corrected regional climate change
projections. International Journal of Biometeorology
59, 1597-1605. doi:10.1007/s00484-015-0966-1
Kartschall T., Wodinski M., von Bloh W., Oesterle
H., Rachimov C. and Hoppmann D., 2015. Changes
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
420

in phenology and frost risks in Vitis vinifera (cv
Riesling) between 1901 and 2100. Meteorologische
Zeitschrift 24, 189-200. doi:10.1127/metz/2015/0534
Keller M., 2015. The Science of Grapevines. Anatomy
and Physiol ogy. 2nd edition (E lsevi er Academic
Press, London).
Ko tre mba C., Tin tru p G. a nd T rap p M., 2014.
Spätfrostgefährdung des Weinbaugebietes Pfalz - eine
klimatologische und reliefbasierte Betrachtung. In:
Schultz H. R. and Stoll M. (eds.) Deutsches Weinbau
Jahrbuch 2015 (Eugen Ulmer KG, Stuttgart).
Lokys H.L., Junk J. and Krein A. , 2015. Future
changes in human-biometeorological index classes in
th ree re gi ons of Luxem bo urg, Wes te rn-Cent ral
Europe. Advances in Meteorology, Article ID 323856.
doi:10.1155/2015/323856
Lo renz D.H ., E ich hor n K. W., Ble iho lde r H. ,
Klose R., Meier U. and Weber E., 1995. Phenological
growth stages of the grapevine, Vitis vinifera L. ssp.
vinifera. Codes and descriptions according to the
extended BBCH scale. Australian Journal of Grape
and Wine Research 1, 100-103. doi:10.1111/j.1755-
0238.1995.tb00085.x
Martinez-Lüscher J., Morales F., Sanchez-Diaz M.,
Delrot S., Aguirreolea J., Gomes E. and Pascual I.,
2015. Climate change conditions (elevated CO2 and
temperature) and UV-B radiation affect grapevine
(Vi tis vini fer a c v. Temp ran ill o) le af ca rbo n
assimilation, altering fruit ripening rates. Plant
Science 236, 168-176. doi:10.1016/j.plantsci.2015.
04.001
Matzarakis A., Rammelberg J. and Junk J., 2013.
As sessm ent of t he rma l biocli mate and tou rism
climate potential for central Europe – the example of
Luxembourg. Theoretical and Applied Climatology
114, 193-202. doi:10.1007/s00704-013-0835-y
Molitor D., Baus O., Hoffmann L. and Beyer M.,
2016a. Meteo rological cond itions determi ne the
thermal- temporal position of the annual Botrytis
bunch rot epidemic on Vitis vinifera L. cv. Riesling
grapes. Oeno One 50, 231-244. doi:10.20870/oeno-
one.2016.50.3.36
Molitor D., C aff arr a A., Sini goj P ., Pert ot I.,
Hoffmann L. and Junk J., 2014a. Late frost damage
risk for viticulture under future climate conditions: a
case study for the Luxembourgish winegrowing
region. A ustra lian Jou rnal of Gr ap e an d Wi ne
Research 20, 160-168. doi:10.1111/ajgw.12059
Molitor D. , Junk J. , Evers D., Hoffm ann L. and
Beyer M., 2014b. A high resolution cumulativ e
degree day based model to simulate phenological
develop ment of grapevi ne. Americ an Journal of
Enolo gy and Viticulture 65, 72-80. doi:10.5344/
ajev.2013.13066
Molitor D. and Keller M., 2016. Yield of Müller-
Th urg au a nd R ie sli ng g rapev ine s is al tered by
meteorological conditions in the current and previous
growing seasons. Oeno One 50, 245-258. doi:10.
20870/oeno-one.2016.50.4.1071
Molitor D., Schultz M., Mannes R., Pallez-Barthel
M., Hoffmann L. and Beyer M., 2019. Semi-minimal
pruned hedge: a potential climate change adaptation
strategy in viticulture. Agronomy 9, 173. doi:10.3390/
agronomy9040173
Molitor D., Udelhoven T., Ney S., Hoffmann L. and
Pfister L., 2016b. Historical vintage descriptions from
Luxembourg - an indicator for the climatic conditions
in the past? Vitis 55, 23-30. https://ojs.openagrar.de/
index.php/VITIS/article/view/4713
Moncur M.W., Rattigan K., Mackenzie D.H. and Mc
Intyre G.N., 1989. Base temperatures for budbreak
and leaf appearance of grapevines. American Journal
of Enology and Viticulture 40, 21-26.
Mosedale J.R., Wilson R.J. and Maclean I.M.D.,
2015. Climate change and crop exposure to adverse
weather: ch ang es to fro st ris k a nd gra pev ine
fl ow er in g con di ti on s. PLoS One 10, e0141218.
doi:10.1371/journal.pone.0141218
Nakicenovic N. and Swart R., 2000. Special Report
on Emissions Scenarios: A Special Report of Working
Group III of the Intergovernmental Panel on Climate
Change (Cambridge University Press, Cambridge).
https://www.ipcc.ch/report/emissions-scenarios/
Nendel C., 2010. Grapevine bud break prediction for
co ol wi nte r cli mat es. Int ern ati ona l Jou rna l of
Biometeorology 54, 231-241. doi:10.1007/s00484-
009-0274-8
Oliveira M., 1998. Ca lculati on of budbrea k and
flowering base temperatures for Vitis vinifera cv.
Touriga Francesa in the Douro region of Portugal.
American Journal of Enology and Viticulture 49, 74-
78.
Parisi S.G., Antoniazzi M.M., Cola G., Lovat L.,
Mariani L., Morreale G., Zoltan K. and Calo A.,
2014. Sprin g thermal resources for grapevine in
Köszeg (Hungary) deduced from a very long pictorial
time series (1740–2009). Climatic Change 126, 443-
454. doi:10.1007/s10584-014-1220-2
Parker A.K., de Cortazar-Atauri I., van Leeuwen C.
and Chuine I., 2011. General phenological model to
characterise the timing of flowering and veraison of
Vitis vinifera L. Australian Journal of Grape and
Wine Research 17, 206-216. doi:10.1111/j.1755-
0238.2011.00140.x
Pa rke r A.K ., Raw V ., Mar tin D ., H ayc ock S.,
Sherman E. and Trought M.C.T., 2016. Reduced
grapevine canopy size post-flowering via mechanical
trimming alters ripening and yield of ‘Pinot noir’.
Vitis 55, 1-9. https://ojs.openagrar.de/index.php/
VITIS/article/view/6005
Ramos M.C., 2017. Projection of phenology response
to climate change in rainfed vineyards in north-east
© 2019 International Viticulture and Enology Society - IVESOENO One 2019, 3, 409-422 421

Spain. Agricultural and Forest Meteorology 247,
104-115. doi:10.1016/j.agrformet.2017.07.022
Sadras V. and Moran M.A., 2013. Nonlinear effects
of elevated temperature on grapevine phenology.
Agricultural and Forest Meteorology 173, 107-115.
doi:10.1016/j.agrformet.2012.10.003
Schultz H.R., 1992. An empirical mode l for the
sim ulation o f l eaf appe ara nce a nd le af- are a
development of primary shoots of several grapevine
(Vi tis vi nif era L. ) c ano py- sys tem s. Sc ien tia
Ho rti cul tu rae 52, 179-200. doi:10.1016/0304-
4238(92)90020-D
Schultz H.R. and Hofmann M., 2017. The ups and
downs of environmental impact on grapevines: future
challenges in temperate viticulture. In: Geros H.,
Ch ave s M. M ., Gil H . M. and D elr ot S . (e ds. )
Grapevine in a Changing Environment: A Molecular
and Ecophysiological Perspective (John Wiley and
Sons). doi:10.1002/9781118735985.ch2
Sgubin G., Swingedouw D., Dayon G., Garcia de
Co rtazar- Ata uri I ., Olla t N., P agé C . and v an
Leeuwen C., 2018. The risk of tardive frost damage
in Fren ch vi ney ard s i n a chan gin g c lim ate .
Agricultural and Forest Meteorology 250-251, 226-
242. doi:10.1016/j.agrformet.2017.12.253
Stoll M., Bi schoff-Schaefer M., Lafontaine M.,
Tittmann S. and Henschke J., 2013. Impact of various
leaf area modifications on berry maturation in Vitis
vinifera L. Riesling. Acta Horticulturae 978, 293-
299. doi:10.17660/ActaHortic.2013.978.34
Tomasi D., Jones G.V., Gius t M., Lovat L. and
Gaiotti F., 2011. Grapevine phenology and climate
change: relationships and trends in the Veneto region
of Italy for 1964-2009. American Journal of Enology
and Viticulture 62, 329-339. doi:10.5344/ajev.2011.
10108
Tonietto J. and Carbonneau A., 2004. A multicriteria
climatic classification system for grape-growing
regions w orl dwi de. Agri cul tur al and Fore st
Meteorology 124, 81-97. doi:10.1016/j.agrformet.
2003.06.001
Trought M.C.T., Parker A.K. and van Leeuwen C.,
2015. Can a change in vineyard practice mitigate
warming due to climate change? Acta Horticulturae
1082, 397-402. doi:10.17660/ActaHortic.2015.
1082.55
va n Lee uwe n C. and De str ac Irvi ne A., 2017 .
Modified grape composition under climate change
conditions requires adaptations in the vineyard. Oeno
One 51, 147-154. doi:10.20870/oeno-one.2017.51.2.
1647
van Pelt S.C., Beersma J.J., Buishand T.A., van den
Hurk B.J.J.M. and Kabat P., 2012. Future changes in
extreme precipitation in the Rhine basin based on
gl ob al and re gi on al climate mo del simula ti on s.
Hydrology and Earth System Sciences Discussions 9,
6533-6568. doi:10.5194/hessd-9-6533-2012
Webb L.B., Whetton P.H. and Barlow E.W.R., 2007.
Modelled impact of future climate change on the
phenology of winegrapes in Australia. Australian
Journal of Grape and Wine Research 13, 165-175.
doi:10.1111/j.1755-0238.2007.tb00247.x
Zapata D., Salazar M., Chaves B., Keller M. and
Ho oge nb oom G., 2015. Est im ati on of the bas e
temperature and growth phase duration in terms of
th ermal time fo r f our gra pev ine cul tiv ars .
International Journal of Biometeorology 59, 1771-
1781. doi:10.1007/s00484-015-0985-y
Daniel Molitor and Jürgen Junk
© 2019 International Viticulture and Enology Society - IVES OENO One 2019, 3, 409-422
422