Historical phenology: Grape ripening as a past climate indicator

Abstract

French records of grape-harvest dates in Burgundy were used to reconstruct spring-summer temperatures from 1370 to 2003 using a process-based phenology model developed for the Pinot Noir grape. Our results reveal that temperatures as high as those reached in the 1990s have occurred several times in Burgundy since 1370. However, the summer of 2003 appears to have been extraordinary, with temperatures that were probably higher than in any other year since 1370.

Full text

French records of grape-harvest dates in
Burgundy were used to reconstruct
spring–summer temperatures from
1370 to 2003 using a process-based phenol-
ogy model developed for the Pinot Noir
grape. Our results reveal that temperatures
as high as those reached in the 1990s have
occurred several times in Burgundy since
1370. However, the summer of 2003
appears to have been extraordinary, with
temperatures that were probably higher
than in any other year since 1370.
Biological and documentary proxy
records have been widely used to reconstruct
temperature variations to assess the excep-
tional character of recent climate fluctua-
tions1–3.Grape-harvest dates, which are
tightly related to temperature, have been
recorded locally for centuries in many Euro-
pean countries. These dates may therefore
provide one of the longest uninterrupted
series of regional temperature anomalies
(highs and lows) without chronological
uncertainties1.
In Burgundy, these officially decreed
dates have been carefully registered in parish
and municipal archives since at least the early
thirteenth century. We used a corrected and
updated harvest-date series4from Burgundy,
covering the years from 1370 to 2003, to
reconstruct spring–summer temperature
anomalies that had occurred in eastern
France. To convert historical observations
into temperature anomalies, we used a
process-based phenology model for Pinot
Noir, the main variety of grape that has been
continuously grown in Burgundy since at
least the fourteenth century5(for details, see
supplementary information).
Our yearly reconstruction is significantly
correlated (Table 1) with summer tempera-
tures deduced from tree rings in central
brief communications
NATURE
|
VOL 432
|
18 NOVEMBER 2004
|
www.nature.com/nature 289
France6(correlation coefficient, r0.53), the
Burgundy part of a spatial multi-proxy recon-
struction2(r0.69) and observed summer
temperatures in Paris7(r0.75), central
England8(r0.53) and the Alps9(r0.45).
Figure 1 shows two early warm decadal
fluctuations: one in the 1380s (0.72 °C)
and one in the 1420s (0.57 °C), both above
the 95th percentile. The warm period of the
1420s was followed by a cold period that
lasted from the mid-1430s to the end of the
1450s (0.45 °C, under the 10th percentile).
Our series also reveals particularly warm
events, above the 90th percentile, in the
1520s and between the 1630s and the 1680s.
These decades were as warm as the end of the
twentieth century. The high-temperature
event of 1680 was followed by a cooling,
which culminated in the 1750s (under the
5th percentile) — the start of a long cool
period that lasted until the 1970s.
The inferred anomaly for the summer of
2003 represents an unprecedented event. It
was 5.86 °C warmer than the reference
period (1960–89), whereas the next highest
anomaly during the whole period was
4.10 °C in 1523. This confirms and refines
the conclusions of previous studies2,10 about
the exceptional warmth of the 2003 summer
in France.
Grape-harvest dates offer the potential
to trace geographical variations in tempera-
ture over large parts of Europe and the
Middle East over past centuries. This
climate, history and phenology synergy can
be used to reconstruct temperatures that
will substantially add to the long proxy-
record databases and provide insight into
regional-scale climate variations.
Isabelle Chuine*, Pascal Yiou†,Nicolas
Viovy†, Bernard Seguin‡, Valérie Daux†,
Emmanuel Le Roy Ladurie§
Grape ripening as a past climate indicator
Summer temperature variations are reconstructed from harvest dates since 1370.
Year
Temperature anomalies (ºC)
–2.0 –1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0
1400 1450 1500 1550 1600 1650 1700 1750 1800 1850 1900 1950 2000
0510
15
2 (1901 – 2003)
σ
Number of stations
Figure 1 April–August temperature anomalies in Burgundy, France, as reconstructed from grape-harvest dates from 1370 to 2003.
Yearly anomalies are in black and the 30-year gaussian filter is in yellow. Confidence intervals due to vineyard differences, with an
11-year smoothing, are shaded in blue; these are estimated from the inter-station variability upper 90th and lower 10th percentiles, and
are determined when there are more than three available observations in a year. Orange line (number of stations) represents the number
of observed harvest dates for each year, indicating where the confidence intervals are computed. Confidence intervals with two s.e., due
to the regression between observed and reconstructed temperature in Dijon, are in purple. These were obtained by regressing the recon-
structed temperature with the observed temperature over 1880–2000. Green horizontal (zero) line is determined from the 1960–89
reference period. Red horizontal lines represent the 2

interval of the reconstructed temperature for the twentieth century (1901–2003).
Vertical arrows indicate warm decadal periods (red) above the 90th percentile and the cold trends (blue) under the 10th percentile.
A fifteenth-century depiction of the grape harvest from Les Très
Riches Heures du Duc de Berry,a medieval book of hours.
Table 1 Linear correlation coefficients between reconstructed temperatures
Series Tree rings Multi-proxy Multi-proxy Paris Central England Alps
(JJA) (JJA) (AMJJA) (AMJJA) (AMJJA) (JJA)
Time range 1750–1975 1500–1998 1659–1998 1787–2000 1663–1992 1760–1998
Linear correlation 0.530.09 0.570.06 0.690.06 0.750.07 0.530.08 0.450.09
coefficient, r
(grape-harvest date)
Linear correlation coefficients between temperatures reconstructed from grape-harvest dates in Burgundy and other observed or reconstructed temperatures
are shown. Comparison with temperatures reconstructed from a tree-ring database6used the closest grid point to Burgundy in the data set. Comparison with
multi-proxy reconstructed temperatures2used the four closest grid points to Burgundy in the data set. The Paris7,central England8,and Alps9series are
observed temperatures (instrumental series). Correlations were computed on the common time intervals between two time series. 95% confidence intervals
are shown. JJA, June to August; AMJJA, April to August.
V&A MUSEUM/WWW.BRIDGEMAN.CO.UK
18.11 brief comms 289 newnew 15/11/04 10:16 am Page 289
© 2004Nature Publishing Group

*CEFE–CNRS, 1919 route de Mende, 34293
Montpellier, France
†LSCE–CEA–CNRS CE Saclay l’Orme des
Merisiers, 91191 Gif-sur-Yvette, France
e-mail: pascal.yiou@cea.fr
‡INRA Site Agroparc, domaine Saint-Paul, 84914
Avignon Cedex 9, France
§Collège de France, 75231 Paris Cedex 05, France
1. Pfister,C. We tternachhersage. 500 Jahre Klimavariationen und
Naturkatastrophen 1496–1995 (Haupt, Bern, Stuttgart and
Wien, 1999).
2. Luterbacher,J., Dietrich, D., Xoplaki, E., Grosjean, M. &
Wanner, H. Science 303, 1499–1503 (2004).
3. Jones, P. D. & Mann, M.E. Re v. Geophys. 42,
doi:10.1029/2003RG000143 (2004).
4. Le Roy Ladurie, E.Histoire du Climat depuis l’An Mil (Champs
Flammarion, Paris, 1983).
5. Robinson, J.,Dinsmoor, A. & Smart, R. E.The Oxford
Companion to Wine (Oxford University Press, 1999).
6. Briffa, K. R.,Jones, P. D. & Schweingruber, F. H. Quat. Res. 30,
36–52 (1988).
7. Renou, E. Ann. Bur. Centr. Météorol. B 195–226 (1887).
8. Manley,G. Q. J. R. Meteorol. Soc. 100, 389–405 (1974).
9. Boehm, R. et al. Int. J. Climatol. 21, 1779–1801 (2001).
10.Schär, C. et al. Nature 427, 332–336 (2004).
Supplementary information accompanies this communication on
Nature’s website.
Competing financial interests: declared none.
more than 27% of global land area) in
1958–99. For 1950–2000,the trends of g lobal
annual average Tmin for windy, calm and all
conditions were identical (0.190.06 °C per
decade; Fig. 1a). So, urbanization has not
systematically exaggerated the observed
global warming trends in Tmin.The same can
be said for poor instrumental exposure and
microclimatic effects, which are also reduced
when instruments are well ventilated5.
When the criterion for ‘calm’ was
changed to the lightest decile of wind
strength, the global trend in Tmin was
unchanged. The analysis is therefore robust
to the criterion for ‘calm’. To assess the effect
of time differences between the reanalysis4
daily-average winds and Tmin,I repeated
the analysis using 26 stations in North
America and Siberia that have hourly or
six-hourly reports of simultaneous temper-
ature and wind. Again, windy and calm
nights warmed at the same rate, in this case
by 0.20°C per decade.
Because a small sample was used, I com-
pared global trends for the reduced period
1950–93 with published all-conditions
trends for that period based on a sample of
over 5,000 stations6.All differences were
within 0.02 °C per decade. This robustness
arises because of the spatial coherence of sur-
face temperature variations and trends7.
The global annual result conceals a
relative warming of windy nights in winter
in the extratropical Northern Hemisphere
(Fig. 1b), mainly in western Eurasia. The
observed tendency to an increased positive
phase of the North Atlantic Oscillation8
implies that the windier days in western
Eurasia had increased warm advection from
the ocean9,yielding greater warming. In
summer in the extratropical Northern
Hemisphere (Fig. 1c), there was no relative
change in Tmin on windy nights. At that time
of year, atmospheric circulation changes are
less influential, but an urban warming signal
is still absent. In the tropics, calm nights
warmed relative to windy nights on an annu-
al average, but only by 0.020.01 °C per
decade, which is much less than the overall
tropical warming in Tmin (0.160.03 °C).
This analysis demonstrates that urban
warming has not introduced significant biases
into estimates of recent global warming.
The reality and magnitude of global-scale
warming is supported by the near-equality of
temperature trends on windy nights with
trends based on all data.
David E. Parker
Hadley Centre, Meteorological Office,
Exeter EX1 3PB, UK
e-mail: david.parker@metoffice.com
1. Kalnay, E. & Cai, M.Na ture 423, 528–531 (2003).
2. Peterson, T.C.J. Clim. 16, 2941–2959 (2003).
3. Johnson, G. T. et al. Bound. Layer Meteorol. 56, 275–294 (1991).
4. Kalnay, E. et al. Bull. Am. Meteorol. Soc. 77, 437–471 (1996).
5. Parker,D.E.Int. J. Climatol. 14, 1–31 (1994).
6. Easterling, D.R. et al. Scie nce 277, 364–367 (1997).
7. Jones, P. D., Osborn, T. J.& Briffa, K. R. J. Clim. 10, 2548–2568
(1997).
8. Folland, C. K. et al. in Climate Change 2001: The Scientific Basis.
Contribution of Working Group I to the Third Assessment Report
of the Intergovernmental Panel on Climate Change (eds
Houghton,J.T.et al.) 99–181 (Cambridge Univ.Press,
Cambridge, UK, 2001).
9. Hurrell, J.W.& van Loon, H. Climat. Change 36, 301–326
(1997).
10.Digg le,P. J.,Liang, K. Y.& Zeger, S. L. Analysis of Longitudinal
Data (Clarendon, Oxford, 1999).
Competing financial interests: declared none.
brief communications
290 NATURE
|
VOL 432
|
18 NOVEMBER 2004
|
www.nature.com/nature
Climate
Large-scale warming
is not urban
Controversy has persisted1,2 over the
influence of urban warming on
reported large-scale surface-air tem-
perature trends. Urban heat islands occur
mainly at night and are reduced in windy
conditions3.Here we show that, globally,
temperatures over land have risen as much
on windy nights as on calm nights, indicat-
ing that the observed overall warming is not
a consequence of urban development.
Observations of the minimum tempera-
ture (Tmin) over 24 hours at 264 stations
worldwide since 1950 were expressed as
anomalies, relative to the period 1961–90
where possible. Coverage of Tmin data was
good north of 20° N, in Australasia and in
the western tropical Pacific, but poor in
Africa, South America, Antarctica and parts
of southern Asia. Reanalysed4daily-average
near-surface wind components were used
to classify the Tmin anomalies into ‘windy’
(upper tercile) and ‘calm’ (lower tercile)
conditions. Daily average wind speeds were
used because the timings of temperature
extremes are not known. For stations
between 140° E and the dateline, Tmin —
which occurs most frequently in the early
morning — was matched with the previous
day’s speed.This is because the early morn-
ing in terms of universal time (equivalent
to Greenwich Mean Time) is still in the
previous day in the Far East.
Annual and seasonal anomalies of Tmin
were gridded on a 5°5° resolution for
windy, calm and ‘all’ conditions. Coverage
was at least 200 grid boxes (equivalent to
1950
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
–3.0
–2.0
–1.0
0.0
Tmin (ºC)
Year
1.0
2.0
4.0
3.0
1960 1970 1980 1990 2000
a
b
c
Atmospheric science
Early peak in Antarctic
oscillation index
The principal extratropical atmospheric
circulation mode in the Southern
Hemisphere, the Antarctic oscillation
(or Southern Hemisphere annular mode),
represents fluctuations in the strength of
the circumpolar vortex and has shown a
trend towards a positive index in austral
summer in recent decades, which has been
linked to stratospheric ozone depletion1,2
and to increased atmospheric greenhouse-
gas concentrations3,4.Here we reconstruct
the austral summer (December–January)
Antarctic oscillation index from sea-level
pressure measurements over the twentieth
century5and find that large positive values,
and positive trends of a similar magnitude
Figure 1 Anomalies in Tmin for windy (red) and calm (blue)
conditions. a, Annual global data; b, winter data (December to
February) for Northern Hemisphere land north of 20° N; c, sum-
mer data (June to August) for Northern Hemisphere land north of
20° N. The linear trend fits, and the 2

error ranges given in
the text, were estimated by restricted maximum likelihood10,
taking into account autocorrelation in the residuals. As expected
from the reduced stratification of the boundary layer, Tmin is, on
average, warmer on windy nights than on calm nights.
18.11 brief comms 289 newnew 15/11/04 10:16 am Page 290
© 2004Nature Publishing Group