120

0

125

0

130

0

135

0

140

0

145

0

150

0

155

0

160

0

1650

170

0

175

0

180

0

185

0

190

0

University of Bern

Institute of History & Oeschger Centre for Climate Change Research

Länggassstrasse 49

3012 Bern, Switzerland

Unversity of Eastern Finland

Department of Geographical and Historical Studies

P. O. Box 111

80101 Joensuu, Finland

Heli Huhtamaa

Crop failures and extreme climate events in historical Finland- Combining written sources and tree-ring data

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

-3

-2

-1

0

1

2

3

120

0

1225

125

0

127

5

130

0

132

5

135

0

137

5

140

0

142

5

1450

147

5

150

0

152

5

155

0

157

5

160

0

162

5

165

0

167

5

170

0

172

5

175

0

177

5

180

0

182

5

185

0

187

5

190

0

1925

195

0

197

5

200

0

-6

-4

-2

0

2

4

6

8

10

12

14

-5

-4

-3

-2

-1

0

1

2

3

4

5

186

0

186

5

187

0

187

5

188

0

188

5

189

0

189

5

190

0

190

5

191

0

Rye Barley MXD Apr-Aug T

Rye

an

dbarley

yie

ld-t

o-s

eed

ratio

an

dM

XD

-pro

xy (

z-s

core

s) O

bse

rve

dF

inla

nd

April-A

ug

ust

tem

pe

ratu

re(ºC

)

Maximum latewood density (MXD) data:

• Matskovsky & Helama (2014) Testing summer

temperature reconstruction. Climate of the Past

10

• Helama et al. (2014) A palaeo-temperature record

based on tree rings. Geochronometria 41

-3

-2

-1

0

1

2

3

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000Tem

pera

ture

anom

alie

s (

ºC)

wrt

19

61

-1990 m

ean

Fig. 7. Crop failure (orange bar), and MXD proxy (grey line)

Fig. 5. Finland in the 1539 Carta marina.

Combining information from written sources and tree-rings

All of the late-19th century crop failures coincided with years when the MXD

series holds extremely low values (below the 2nd percentile of the period

1200–2000). Thus, the historical yield statistics, early meteorological

observations and the MXD data suggests that crop failures were connected

to cold climate extremes. However, historical records of large-scale crop

failures prior the mid-19th century can be found on years, when the MXD

series do not hold extremely low values (Figure 7).

The lack of correspondence between the MXD series and the historical

crop-failure chronology might be due to:

1. Nature of the crop failure data prior the 19th century

a. Inaccurate dating of the crop failure events (especially prior the 17th

century)

b. Errors considering the extent of the crop failure. Local poor harvest

might have been marked as countrywide crop failures (due to lack of

information or economic interests, such as tax reliefs).

2. Nature of the MXD series. For example, the sample size of the

southern Finland MXD series is smaller than 10 prior the 1200s.

3. Human component

Human actions could have increased the sensitivity for crop failures.

Therefore, not only climate extremes, but also minor disturbances may

have had severe consequences. In the case of historical Finland, at least

four key factors should be considered:

b. Wars and social distress. Military recruiting took labour off from the

fields, and during wartime taxes tend to increase. In addition, the

peasants were obligated to feed the armies passing by.

c. Taxation. Taxes and other payments were partly paid in grain till the late

19th century. Fixed tax products, prices and practices burdened the cycle

of crop cultivation.

d. Opportunities and regulations. Were peasants and institutions able to

cope with climate extremes? For example, could peasants clear new

fields for cultivation? Or, did the state offer loans for seed grain and

ease the tax burden on years of poor harvest?

Introduction

• Climatic conditions have largely influenced agricultural

practices and crop variations in Finland.

• Short growing season, and related spring and summer

temperatures, were the main limiting factors for the old

agriculture.

• The aim is to examine written primary sources and tree-ring

studies in order to:

1. construct a chronology of large-scale crop failures,

2. investigate on what frequency extreme climate events

contributed to crop failures in 1300–1900 Finland.

An analogue from the 19th century

Quantifying the past effects of climate and weather on agricultural

production is challenging, as evidence for past crop failures in Finland is

scant, especially prior to the 16th century. The existing information remains

both temporally and spatially fragmented (Table 1). However, reliable crop

yield statistics, meteorological observations and other socio-environmental

data with high accuracy and precision are available from the mid-19th century

onwards. Therefore, the example of 19th century crop yield responses to

climate and weather fluctuations is used as an analogue for the earlier

times.

The contemporaries named early autumn night-frost as the main

reason for large-scale crop failures. Yet, the consequences of night-frost on

crop yields were partly dictated by preceding spring and early summer

temperatures.

Low spring temperatures delayed the onset of the growing season,

sowings and flowering. Hence, the ripening extended to late summer or

early-autumn, when the risk of crop failure due to night-frost was increased.

Therefore, expectedly, the 19th century data of crop yields correlated

positively with warm season (April–August) monthly-mean temperatures. On

years of countrywide poor harvest and crop failure due to night-frost (1862,

1867, 1877, 1892 and 1902), the April–August mean temperature was 1.5 °C

lower (or more) than the late-19th century mean.

Continuous meteorological observations started in Finland 1829.

Where and when instrumental observations are not available, climate

proxies provide evidence of past climatic fluctuations. In Finland, the highest

paleoclimatic correlations are typically derived from latewood maximum

density (MXD) series (Fig. 1-3). The MXD is used to reconstruct warm

season temperatures. In addition, the MXD data can offer a promising, new

type of material for studying crop failure history (Fig. 2). The MXD data and

the late-19th century crop yield series have a strong correspondence. Thus,

in the case of Finland, the tree-ring materials do not only provide evidence

of past temperature fluctuations, but also surrogate data for past harvest

fluctuations.

Fig. 4. Averaged northern and southern Finland MXD series, observed growing

season temperature and crop yield statistics from 19th Finland

Source: Huhtamaa et al. (2015) Crop yield responses to temperature fluctuations in

19th century Finland. Boreal Environmental Research 40

Archaeology,

Swedish and Russian sources

Fragmentary information (e.g.

letters, manorial accounts)

First documentation of annual

harvest fluctuations, e.g.

tithes, peasant complaints

National statistics

• Annual yield data

• Meteorol.

measurements

Non-continuous & non-

systematic harvest

reports, diaries, etc.

Table 1. Historical sources on harvest and weather fluctuations

Sources

Temporal

precision Annual to daily Monthly to hourlyAnnual to monthlyDecadal to annual

Tree-ring series (annual, local to regional)

Spatial

coverageWhole Finland

Southern and western Finland,

biggest townsSome townsCouple towns

Most parts of the

country

Fig. 1-3 (from top left clockwise). The sampling sites of the MXD series, the

MXD based growing season temperature reconstruction and field correlations

with observed April-August temperatures 1850-2000.

Fig. 6. Parish records of the crop failure in

1601. Sown rye: 210.5 barrels; harvested

rye: 175 barrels, 25 cappas.

(Poster illustrations from Carta Marina (1539)

and Historia de gentibus septentrionalibus

(1555) by Olaus Magnus

Avera

ged n

ort

hern

and s

outh

ern

Fin

land

MX

D s

eries (

z-s

core

s)

Multi-centennial (archaeology) to

annual (chronicles)

a. Agro-ecosystem. The agricultural

productivity was much lower in the

medieval and early modern times.

For example, the yield-to-seed ratio

was in the 17th century 4, whereas

in the 19th century it was 6 (and

today between 30 and 60,

depending on the crop species).

The chronology of past large-scale (1/3 of the provinces or more) crop failures have been gathered from medieval chronicles and letters (prior 16th C), tithe records (16th–17th C), peasant complaints (from 17th C) and national yield statistics (from 19th C).

Conclusions

• Constructing reliable chronology of crop failures prior 16th

century is impossible due to lack of written sources.

• Tree-ring data help to identify when crop failure was caused

by climatic factors.

• In the later half of the 19th century cold April–August

anomalies always contributed to the crop failures.

• Prior the 19th century man made factors increased the crop

production sensitivity.

• Cold summer anomalies caused only minority of the crop failures.

• Yet, in further research, “climate extreme” should be defined time-

specifically, considering the changing socio-environmental factors.

• The comparison between the written records and tree-ring

material suggests that the historical records hold many dating

errors and false events.

2nd percentile

5th percentile

Swiss Climate Summer School: Extreme Events and Climate. Monte Verità, Switzerland, August 23– 28, 2015