A comprehensive study of floods in Poland in the 17th-18th centuries

Abstract

Study region: Poland, with the regions of Baltic Coast and Pomerania, Masuria-Podlasie, Greater Poland, Masovia, Silesia, and Lesser Poland located in the basins of the Baltic Coast rivers, the Vistula River and the Oder River. Study focus: This study focused on completing the documentation of historical floods in Poland before the 19th century and providing a valuable source for historical hydrology studies in Europe. To this end, a comprehensive database of all floods for the 17th-18th centuries was used, that was based on documentary evidence from 293 sources and 978 weather notes describing all flood occurrences. New hydrological insights for the region: The finding of this study revealed the occurrences of 678 floods, including 37 new cases that have been discovered and documented only in this research. Spatial analysis of the results revealed that most of the floods occurred in the Vistula River basin. The number of floods by season was greatest for summer (JJA) (47 %) and smallest for autumn (7 %). Investigation of the origin of floods indicated that rain was the main factor contributing to occurrences of floods in Poland (38 %). The estimation of the intensity of floods showed that most of the floods were "smaller, regional floods" (257 cases) based on the Brázdil et al. (2006b) classification and "extraordinary" (501 cases) in the Barriendos & Coeur (2004) classification.

Full text

Journal of Hydrology: Regional Studies 53 (2024) 101796
Available online 24 April 2024
2214-5818/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
A comprehensive study of oods in Poland in the
17th–18th centuries
Babak Ghazi
a
,
*
, Rajmund Przybylak
a
,
e
, Piotr Oli´
nski
b
,
e
, Michał Targowski
c
,
e
,
Janusz Filipiak
d
, Aleksandra Pospieszy´
nska
a
,
e
a
Department of Meteorology and Climatology, Faculty of Earth Sciences and Spatial Management, Nicolaus Copernicus University, Toru´
n, Poland
b
Department of Medieval History and Auxiliary Sciences of History, Faculty of Historical Sciences, Nicolaus Copernicus University, Toru´
n, Poland
c
Department of Early Modern History and Source Editing, Faculty of Historical Sciences, Nicolaus Copernicus University, Toru´
n, Poland
d
Department of Physical Oceanography and Climate Research, Faculty of Oceanography and Geography, University of Gda´
nsk, Gda´
nsk, Poland
e
Centre for Climate Change Research, Nicolaus Copernicus University, Toru´
n, Poland
ARTICLE INFO
Keywords:
Historical hydrology
Historical oods
Historical climatology
Poland
Central Europe
17th–18th centuries
ABSTRACT
Study region: Poland, with the regions of Baltic Coast and Pomerania, Masuria-Podlasie, Greater
Poland, Masovia, Silesia, and Lesser Poland located in the basins of the Baltic Coast rivers, the
Vistula River and the Oder River.
Study focus: This study focused on completing the documentation of historical oods in Poland
before the 19th century and providing a valuable source for historical hydrology studies in
Europe. To this end, a comprehensive database of all oods for the 17th–18th centuries was used,
that was based on documentary evidence from 293 sources and 978 weather notes describing all
ood occurrences.
New hydrological insights for the region: The nding of this study revealed the occurrences of 678
oods, including 37 new cases that have been discovered and documented only in this research.
Spatial analysis of the results revealed that most of the oods occurred in the Vistula River basin.
The number of oods by season was greatest for summer (JJA) (47 %) and smallest for autumn
(7 %). Investigation of the origin of oods indicated that rain was the main factor contributing to
occurrences of oods in Poland (38 %). The estimation of the intensity of oods showed that most
of the oods were “smaller, regional oods” (257 cases) based on the Br´
azdil et al. (2006b)
classication and “extraordinary” (501 cases) in the Barriendos & Coeur (2004) classication.
1. Introduction
Climate change has increased the public and scientic communities’ concerns about the increasing frequency of oods in river
basins globally (Rohde, 2023; Rottler et al., 2023). Floods are among the most frequent natural disasters. In terms of various economic
and human losses, oods have the most signicant catastrophic impacts on human societies and the environment. Europe has
experienced numerous devastating oods in the last few decades, resulting in considerable economic losses (Bl¨
oschl et al., 2019).
Therefore, to reduce ood hazard risk, long-term comprehensive studies are necessary (Brunner et al., 2020; Merz et al., 2021). To this
aim, it is crucial to uncover the intensity and genesis of oods in past centuries (Hundecha et al., 2020). Considering the limited time
* Corresponding author.
E-mail address: babak.ghazi@doktorant.umk.pl (B. Ghazi).
Contents lists available at ScienceDirect
Journal of Hydrology: Regional Studies
journal homepage: www.elsevier.com/locate/ejrh
https://doi.org/10.1016/j.ejrh.2024.101796
Received 25 January 2024; Received in revised form 28 March 2024; Accepted 20 April 2024

Journal of Hydrology: Regional Studies 53 (2024) 101796
2
frame of instrumental hydrological observations, historical hydrology occupies a position at the interface between hydrology and
history (Br´
azdil et al., 2006b). Reconstruction of historical oods can provide a better understanding of ood hazards over the
long-term perspective and improve the ability to predict future events (Bl¨
oschl et al., 2017; Wilhelm et al., 2019). Multiple sources
from historical hydrology provide useful information about climate variability over past centuries that predate the availability of
instrumental records (García-Barr´
on et al., 2023). Documentary evidence constitutes a reliable form of proxy data in terms of
providing credible information about temporal and spatial trends of past climatic and hydrological events (i.e., oods and droughts)
(Kjeldsen et al., 2014). In recent decades, European ood records have been reconstructed using diverse documentary evidence (Cœur
and Lang, 2008; Rets¨
o, 2015; Bl¨
oschl et al., 2020; García-Barr´
on et al., 2023). Studies analyzing the history of oods in Poland in the
last millennium, in particular before the 19th century are written almost exclusively in the Polish language and are usually limited to
general information about the year or region(s) of oods. Moreover, majority of these studies are focused on the last 200 years.
Cyberski et al. (2006) used documentary evidence to review the history of oods on the Vistula River from AD 988–2006. The
authors derived historical oods for the Upper Vistula, Middle Vistula and Lower Vistula from documentary resources in the
pre-instrumental period. The results showed that the rivers in the Carpathian tributaries (Soła, Dunajec and Skawa Rivers) were the
main contributors to ood events on the Upper Vistula. On average, the Soła River made the greatest contribution to ooding of the
Upper Vistula connected with the topography, geology and precipitation conditions of drainage basins. The history of oods for the last
millennium limited to Upper Vistula was also presented by Nachlik and Kundzewicz (2016). A study of oods for period 1799–2001
presented by Fal and Dąbrowski (2001) revealed that 64 % of them in the Middle Vistula were caused by snowmelt. It is reported that,
on the Middle Vistula, snowmelt-induced ood events happened mainly in March, April and February, with these months accounting
for 44 %, 28 % and 16 %, respectively, while rain-induced oods were most common in July (29 %), August (24 %) and June (20 %).
On the other hand, the majority of historical oods in the Lower Vistula occurred due to their freezing in winter and the break-up of ice
jams in spring (Grze´
s, 1991; Embleton and Embleton-Hamann, 1997). Although these studies provide valuable information about the
history of oods in Poland, they lack detailed information such as description of events and the intensity and primary cause of oods.
Unlike for Poland, for other European countries such as Austria, Czech Republic, Hungary, Germany, Sweden and Switzerland there
exist previous studies that have assessed historical oods (Pster, 2002; Br´
azdil et al., 2005b, 2006a, 2011; Rohr, 2006, 2007; Rets¨
o,
2015).
For the reasons described above, the scientic literature about oods in Poland, including in historical times, is rarely known by the
non-Polish scientic community. To ll this knowledge gap, we have already created two comprehensive databases – one for the
11th–15th centuries (https://doi.org/10.18150/WD18XJ) and one for the 16th century (https://doi.org/10.18150/T3RXRI), and we
have reconstructed the history of oods in Poland before the 17th century (Ghazi et al., 2023a, 2023b). However, there is still no
comprehensive study of oods in the 17th–18th centuries in Poland. The aim of this study is to show the results of our work focusing on
ood reconstructions for Poland for the 17th–18th centuries. As a result, the entire history of ood occurrences in Poland will be
known for the pre-instrumental period in hydrology, which started in Europe (including Poland) usually at the beginning of the 19th
century (Br´
azdil et al., 2006b). The results of this and two previous studies (Ghazi et al., 2023a, b) improve knowledge on the historical
climatology and hydrology of Central Europe.
Fig. 1. (a) Changes to the borders of Poland in historical times (source: https://commons.wikimedia.org/wiki/File:Border_changes_in_history_of_
Poland.png) (b), Geographical location of contemporary Poland and main river basins.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
3
Table 1
Years of ood occurrences in Poland for the 17th–18th centuries.
Current study (all of
Poland)
(Grze´
s 2008) (all
of Poland)
Majewski (1993)
(Vistula Delta only)
Biela´
nski
(1997) (Upper
Vistula River
only)
Mudelsee et al.
(2003) (Oder River
basin)
Kasprzak (2010)
(Oder River basin)
(Bl¨
oschl et al. 2020)
Middle
Oder
River
Vistula
River
1601 (2), 1602 (4),
1603 (2), 1604
(4), 1605 (4),
1606 (4), 1607
(2), 1608 (4),
1609 (4), 1610,
1611 (2), 1612
(6), 1613 (3),
1614 (4), 1616
(2), 1621 (6),
1622 (8), 1623,
1624 (8), 1625
(3), 1628 (6),
1629, 1630 (2),
1631, 1633 (2),
1634, 1635 (7),
1636, 1638,
1639 (3), 1644,
1645, 1646 (4),
1647 (3), 1648
(2), 1649 (5),
1650 (11), 1651
(9), 1652 (7),
1653, 1654,
1655 (6), 1656
(4), 1657 (3),
1658 (4), 1659
(4), 1660 (4),
1661 (5), 1662
(10), 1663 (9),
1665 (5), 1666
(3), 1667 (3),
1668 (4), 1669
(2), 1670 (5),
1671 (6), 1672
(5), 1673 (7),
1674 (7), 1675
(18), 1676 (2),
1677 (3), 1678,
1679 (3), 1680
(4), 1685 (3),
1686 (2), 1687
(2), 1688 (6),
1689 (9), 1690
(3), 1691, 1692
(6), 1693 (5),
1694 (4), 1695
(3), 1696 (3),
1697, 1698 (10),
1699 (2), 1700
(2), 1701 (2),
1702 (13), 1703
(17), 1706,
1707, 1708 (3),
1709 (12), 1710
(2), 1711 (2),
1712 (7), 1713
(18), 1714 (9),
1715 (10), 1716,
1717 (5), 1718
(8), 1719 (3),
1720 (2), 1721
(4), 1723 (4),
1724, 1725 (2),
1729 (7), 1730
1601, 1602, 1605,
1606, 1607, 1611,
1612, 1613, 1622,
1624, 1635, 1639,
1651, 1652, 1657,
1661, 1666, 1668,
1669, 1670,
1672, 1673, 1674,
1675, 1676, 1677,
1679, 1680, 1689,
1693, 1700, 1706,
1709, 1713, 1716,
1717, 1718, 1719,
1720, 1721, 1724,
1729, 1731, 1736,
1737, 1738, 1741,
1742, 1743, 1744,
1745, 1749, 1750,
1753, 1757, 1761,
1764, 1765, 1767,
1774, 1778, 1780,
1782, 1783, 1784,
1785, 1786, 1789,
1792, 1794
1602, 1606, 1613,
1621, 1624, 1633,
1639, 1646, 1651,
1652, 1567, 1658,
1659, 1660, 1661,
1662, 1663, 1668,
1669, 1671, 1672,
1674, 1675, 1676,
1677, 1679, 1680,
1693, 1695, 1708,
1713, 1714, 1717,
1718, 1719, 1731,
1734, 1739, 1745,
1749, 1765, 1774,
1775, 1784, 1786,
1605, 1621,
1650, 1652,
1655, 1656,
1662, 1670,
1671, 1687,
1697, 1736,
1774, 1775
1605, 1606, 1612,
1613, 1614, 1616,
1621, 1624, 1625,
1628, 1630, 1633,
1634, 1644, 1646,
1648, 1651, 1652,
1654, 1655, 1658,
1659, 1662, 1663,
1664, 1665 (2),
1667, 1672, 1674,
1675, 1679, 1680,
1685, 1687, 1688,
1689, 1692, 1693,
1694, 1695, 1696,
1698, 1699, 1706,
1708, 1709, 1712,
1713, 1715, 1718,
1729, 1730, 1731,
1736 (2), 1737 (3),
1749, 1751, 1770,
1771, 1775, 1779,
1780 (2), 1783,
1784, 1785, 1789,
1794, 1798, 1799
1604, 1609, 1613,
1649, 1650, 1652,
1653, 1654, 1667,
1685, 1686, 1691,
1694, 1701, 1702,
1703, 1708, 1709,
1710, 1711, 1713,
1714, 1723, 1729,
1736, 1751, 1759,
1766, 1767, 1773,
1778, 1780, 1783,
1785, 1786, 1792,
1795 (2), 1797
1622
1628
1652
1685
1709
1710
1711
1712
1713
1732
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1785
1605
1621
1649
1650
(2)
1651
1652
1655
1656
(2)
1657
1658
1659
1661
(2)
1662
(4)
1663
(3)
1665
1667
1668
1670
1671
1675
(2)
1675
1687
1688
1690
1695
1697
1736
1772
(2)
1774
(3)
1775
1785
(2)
1786
1787
(2)
1795
(continued on next page)
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
4
2. Study area, materials and methods
The borders of Poland have undergone various changes throughout history. In past centuries, the state borders of Poland have
differed from those of the contemporary era (see Fig. 1a). Therefore, to conduct a comprehensive study for assessment of ood events
in Poland, we used the current boundary of this country. The current area of Poland is divided into six main regions (Fig. 1b): Baltic
Coast and Pomerania, Greater Poland, Masovia, Masuria-Podlasie, Lesser Poland and Silesia. These six main regions are parts of three
main river basins in Poland: those of the Vistula River, the Oder River and the Baltic Coast rivers. Thus, we identied all ood events
based on these six main regions and distinguished the main river basin. In a few cases, due to the lack of information about the region
(s) where oods occurred, the miscellaneous category “Poland” was used to describe the area of ood occurrences.
For comparison purposes, the methodology in this research is identical to that which we used in our previous studies analyzing
ood occurrences in Poland in the period 1001–1600 (Ghazi et al., 2023a, b). To summarize, the following procedure was utilized: 1)
construction of a comprehensive database (https://doi.org/10.18150/VLPAFG) of all oods for the 17th–18th centuries, including
information about region(s), place, river basin, date of event (year, month and day if available), weather notes, sources, source quality
and indexation of intensity and origins of ood, 2) evaluation of sources based on their quality (1 – weak, 2 – moderate, 3 – high) to
select best source(s) and weather note(s) for ood events in order to use them for individual classication purposes, 3) independent
classication of every ood made by two historians, two climatologists and one hydrologist, 4) arrangement of the nal classication
in a team discussion, 5) comprehensive statistical analysis of obtained results.
The classication of sources is as follows: 1 – weak (when the information was taken from secondary literature instead of the
original source); 2 – moderate (when the information is written from other centuries than when the ood occurred or the description of
the event differed from that contained in a note written contemporaneously with the event); 3 – high (when information was written in
a source contemporaneous to the ood event and it is precise with exact information about the event). In this research, we used 293
sources (for details see Table S1) and 978 weather notes describing all ood occurrences (https://doi.org/10.18150/VLPAFG).
Table 1 (continued )
Current study (all of
Poland)
(Grze´
s 2008) (all
of Poland)
Majewski (1993)
(Vistula Delta only)
Biela´
nski
(1997) (Upper
Vistula River
only)
Mudelsee et al.
(2003) (Oder River
basin)
Kasprzak (2010)
(Oder River basin)
(Bl¨
oschl et al. 2020)
Middle
Oder
River
Vistula
River
(2), 1731(9),
1732 (3), 1734
(4),
1735 (6), 1736
(26), 1737 (7),
1738, 1739,
1740 (5), 1741,
1742 (2), 1743
(2), 1744 (5),
1745 (5), 1747,
1748 (5), 1749
(3), 1750 (4),
1751 (5), 1752
(2), 1753 (2),
1754, 1755 (2),
1757, 1759,
1761 (2), 1763,
1764, 1765 (5),
1766, 1767 (2),
1668, 1769,
1770 (2), 1771,
1772 (5), 1773
(3), 1774 (12),
1775 (10), 1778,
1779, 1780 (8),
1781, 1782,
1783 (7), 1784
(2), 1785 (10),
1786 (6), 1787
(4), 1788, 1789
(2), 1790 (2),
1792 (2), 1794,
1795 (4), 1797,
1798, 1799
Explanations:
() – more than one ood in the year is given
Bold font – new oods not present in other previous studies
Italic font – ood was classied as “high-water” in our study
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
5
However, for classication of oods’ intensity and origin, we used 700 of the weather notes that are the most reliable and/or describe
oods in greater detail (Table S2). As results from this documentation, various kinds of sources, both published (including original
sources and secondary literature) and unpublished (handwritten, databases), have been used to assess historical oods. The quality of
these sources was evaluated by historians using critical source analysis. Further, the most reliable sources were selected in this pro-
cedure for classication of the intensity and origin of oods.
For classication of origin of oods, we used the method given by Lambor (1954), while their intensity was estimated using criteria
given by (Barriendos and Coeur, 2004; Br´
azdil et al., 2006b)). According to Lambor’s (1954) classication, the primary causes of oods
in Poland are rain (torrential and frontal rain), snowmelt, winter (ice jam and shuga), and storms. Barriendos and Coeur (2004)
proposed ood classication according to the destructive impact, intensity and scale of events. In their classication, oods were
categorized into ordinary, extraordinary and catastrophic oods. The classication of oods proposed by Br´
azdil et al. (2006b) is
slightly modied from that originally introduced by Sturm et al. (2001). Four categories were distinguished; 0 – a ood; 1 – a small,
regional ood; 2 – an above-average or super-regional ood; and 3 – a disastrous ood on a large scale (see Table S3–S4). The intensity
(size) estimation of oods based on documentary evidence is usually very challenging because the available information has a
qualitative character. For this reason, we used the two mentioned classications to overcome the possible uncertainties in addressing
the intensity of oods. These classications have been successfully used and approved both in our previous studies for Poland (Ghazi
et al., 2023a, 2023b) and in other studies analyzing ood occurrences in different parts of Europe (Glaser et al., 2010; Bhat et al., 2019;
Benito et al., 2021).
In the ood database, in some cases, there are several oods in the same year. In that case, when there is clear information in the
weather note(s) about ood occurrences in various rivers, we assume them to be separate oods. On the other hand, when oods
occurred in the same year but in different months or seasons on the same river, we distinguished those separated by an interval greater
than 7 days to have been separate oods.
To distinguish the “ood” and “high-water” events, in the classication of the intensity of oods, we assumed an event was a “high-
water” if it was classied as category “0” (ordinary ood) using the Barriendos and Coeur (2004) classication (see Table S3).
3. Results
A comprehensive assessment of oods in Poland in the 17th–18th centuries demonstrated that, overall, 678 oods occurred
(Table 1 and S5, Fig. 2). Most of them occurred in the Vistula River (349) and Oder River (321) basins, while only a few were recorded
in the Baltic Coast rivers basin (8). In the study period, the highest number of oods (>10 cases) occurred in the following years; 1650
(11), 1675 (18), 1702 (13), 1703 (17), 1709 (12), 1713 (18), 1736 (26) and 1774 (12). Except for the last year, all occurred in a ood-
rich period that we locate between 1650 and 1740 (Fig. 2a). In this time, as many as 417 (61%) oods were registered in the historical
sources. It is interesting to note that the distinguished ood-rich period in Poland contains a relatively “dry” interval in the Vistula
Fig. 2. Number of oods in Poland, AD 1601–1800: (a), All river basins in Poland, (b) Vistula River basin, (c) Oder River basin, (d) Baltic Coast
rivers basin.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
6
River basin; thus, the oods within this period were most frequent in the periods 1650–80 and 1710–40. On the other hand, in the Oder
River basin the ood-rich period started later (in 1690) than it did on average in Poland, but ended at the same time (1740) (Fig. 2b, c).
An increase in ood occurrences in the Vistula River basin is also seen in the period 1775–90.
To show the spatial variability of oods recorded in Poland in the 17th–18th centuries, their frequencies are presented also for its
six main historical-geographical regions (Fig. 3). In general, the most oods were recorded in the Silesia and Baltic Coast and Pom-
erania regions which account for 246 and 166 oods, respectively. The numbers of oods in Lesser Poland (145) and Greater Poland
(69) were also relatively high, while the numbers of oods were small in Masovia (38) and especially small in Masuria-Podlasie (6). For
only eight oods, did we not nd any detailed information about the place/regions of their occurrence (category “Poland”). The year-
to-year course of frequency of ood occurrences stratied into historical-geographical regions is shown in Fig. 4.
Based on the results presented in Fig. 4, ood-rich periods are clearly seen mainly in three regions (Baltic Coast and Pomerania,
Silesia, and Lesser Poland). In the Baltic Coast and Pomerania region, two such periods can be distinguished (ca 1660–1680 and
1710–1745) when 35 (21 %) and 57 (34 %) oods occurred, respectively. In Silesia region, oods were most frequent in the period
from about 1690–1745 (113, 46 %). Two ood-rich periods (ca 1650–1680 and 1775–1790) are also present in the Lesser Poland
region. It is important to note that the timings of these regional ood-rich periods show good correspondence with the distinguished
ood-rich period for the whole of Poland (compare Figs. 2a and 4b, e and f).
In the assessment of the database of registered oods in Poland in the 17th–18th centuries, for almost 76 % of the records, we
identied the month that oods occurred. The analysis of the frequency of oods occurrences in various months of the year
demonstrated that the most occurred in July (22 %), while they were fewest in October (0.5%) (Fig. 5). In general, a bimodal dis-
tribution can be noted, with maxima of ood occurrences in the periods Jun–Aug (primary) and Jan–Apr (secondary). Therefore, in
terms of seasons, the predominant period of oods in Poland in the study period was summer (JJA) (47 %). The frequencies of oods in
spring (MAM) and winter (DJF) were each about half that, accounting for 26 % and 21 %, respectively. Decidedly, the smallest number
of oods was registered for the autumn (SON) (6 %).
Barriendos and Coeur (2004) distinguished category “0” (ordinary oods), which they described as: ‘In-bank ow, no damage,
water discharge can increase but without overowing’ (see Table S3). Such cases were included to our database but were not labeled as
oods (see Table S6). The occurrence of high-water levels in rivers in Poland in the study period reached 43 cases. Most were noted in
the Oder River basin (32). On the other hand, there times fewer high-water levels (11) were registered in the Vistula River basin.
Results of the assessment of the intensity and origin of oods based on the classications mentioned in the previous subsection, are
presented in Fig. 6. According to the Br´
azdil et al. (2006b) classication of intensity of oods, the most common categories were
“Smaller, regional ood” with 257 oods and “Above-average, or supra-regional ood” with 209 cases. About two to three times less
frequent were oods described as “Above-average, or supra-regional ood on a disastrous scale” (119 cases) and “ood” (93). Based on
the Barriendos and Coeur (2004) classication, most of the oods were attributed to the “extraordinary” category (501 cases). Quite a
large number of “catastrophic” category oods was also noted (167 cases). Only ten records belong to the “ordinary” category based on
this classication, which describes not a true ood, but a high water level.
Results of the assessment of the origin of oods using the Lambor (1954) classication revealed that rain was the main contributing
factor for the most occurrences of oods in Poland in the 17th–18th centuries (Fig. 6). Overall, 275 oods were caused by rain, of
which, 64 were related to torrential rain, 3 to frontal rain and 46 to long-lasting, territorially widespread rain. In 162 cases, there is no
detail about rain sub-types. The second most important reason for oods was the occurrence of ice jams on rivers. We found also that
ve oods in Poland in the study region were caused by intentional human activity (usually the destruction of the embankments).
Therefore, we slightly modied the Lambor (1954) classication of origin of oods (see Table S4).
4. Discussion
Retrieving historical oods from documentary evidence, is a challenging task in historical hydrology research due to the various
uncertainties and limitations. These uncertainties in documentary evidence are mainly related to their quality and to human
subjectivity in describing the historical extreme events (Br´
azdil et al., 2005c). However, to provide a comprehensive assessment oods
in Poland in the 17th–18th centuries, we conducted this research based on documentary evidence that had already been successfully
Fig. 3. Number of oods in Poland in the 17th–18th centuries in the six-main regions and miscellaneous category “Poland”.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
7
used in our previous studies for historical oods in Poland (Ghazi et al., 2023a, b).
To assess the results in detail, we compared ndings of this study with available studies for historical oods in Poland (Table 1).
Among all the available studies, only Grze´
s (2008) provided a list of oods for Poland in the 10th–19th centuries. However, in his
work, there is no detailed information about the intensity, origin or documentary references for these events. Majewski (1993) and
Biela´
nski (1997) provided information about historical oods in Poland for the Vistula Delta and Upper Vistula River, respectively. In
Fig. 4. Number of oods in the 17th–18th centuries, (a) miscellaneous category “Poland” (b) Baltic Coast and Pomerania, (c) Greater Poland, (d)
Masovia, (e) Silesia and (f) Lesser Poland. Note: Due to the small number of oods in Masuria-Podlasie region, this region not shown (see Fig. 3).
Fig. 5. Frequency of oods (a) by month and (b) by season.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
8
addition, Mudelsee et al. (2003) studied ood occurrences in Central Europe, and Kasprzak (2010) investigated historical oods in the
Oder River basin. We also extracted a list of oods for Poland based on a study analyzing oods in Europe in the last 500 years (Bl¨
oschl
et al., 2020). Comparing the results of this study with the publications mentioned above showed that the number of recorded oods in
this study (678 for all of Poland, 349 in Vistula River basin, and 321 in the Oder River basin) is signicantly higher than those given by
Grze´
s (2008) (70), Majewski (1993) (45), Biela´
nski (1997) (14), Mudelsee et al. (2003) (71), Kasprzak (2010) (39) and (Bl¨
oschl et al.,
2020) (74) (Table 1). In the comparison of ood occurrences, one ood in 1664 in the Mudelsee et al. (2003) source was classied as a
high water in our database (see Table S6).
In addition, we compared the frequency of oods in the 17th–18th centuries in Poland with available sources (Glaser and Stangl,
2003; Mudelsee et al., 2003, 2006; Br´
azdil et al., 2005a, 2011; Br´
azdil and Kirchner, 2007; Rohr, 2007; Hohensinner, 2015; Rets¨
o,
2015; Bl¨
oschl et al., 2020) from neighboring countries (i.e., Austria, Czech Republic, Germany and Sweden) (Table S7)) to provide an
estimation for ood frequency in Europe during the 17th–18th centuries. However, the reader should be aware of possible biases
related to differences between these countries in terms of the availability of detailed information allowing for the reconstruction of
oods in rivers (including the number of rivers for which information exists). For example, we extracted data for ood occurrences for
ten rivers in Germany, seven rivers in Czech Republic and signicantly fewer for Sweden (4) and particularly for Austria (2) (see
Table S7 for more details).
The results depicted in Fig. 7 show that the 20-year frequency of oods was greater in Poland and Germany than in other Central
Europe countries. Only in two periods (1661–1680 and 1721–1740) was the number of oods higher in Poland than in Germany. This
Fig. 6. Number of oods in Poland in period 1601–1800 based on the classications of: (a) Br´
azdil et al. (2006b), (b) Barriendos and Coeur (2004),
and (c) Lambor (1954), see Table S3 and S4 for classication details. Key: N/A – information not available.
Fig. 7. Frequency of oods in the 17th–18th centuries in Central Europe.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
9
is partly related to the differences in the numbers of rivers and areas of the countries, as well as to differences in availability of sources.
Statistical evaluations demonstrated that ~63 % of oods in Germany and 35 % in the Czech Republic were also observed in Poland.
The corresponding gures for Sweden and Austria were lower, standing at 19 % and 7 %, respectively. This is related to the number of
sources available or used for the reconstruction of historical oods in these countries. In addition, the evidently smaller number of
historical oods in Sweden than in other Central European countries also can be partly related to Sweden’s climate, which is inuenced
by the presence of the Scandinavian mountains (rain shadow).
To obtain a more reliable comparison between oods occurring on the Vistula and Oder rivers (Poland) and in the other analyzed
countries, we chose eight rivers from Table S7 for which there exist great numbers of oods. The number of oods occurring in the
17th–18th centuries in all rivers is presented in Fig. 8. Analysis of results illustrated in Fig. 8 shows that the rivers most prone to oods
in the study period in Central Europe were the Danube, Main and Vistula Rivers (140–160 oods), and those least vulnerable were the
Oder and Vltava Rivers (~60). Quite a large number of oods (>120) also occurred in Elbe River. In Sweden (the Norrstrom River),
oods were registered only 31 times, i.e. usually 2–5 times less frequently than in all Central European rivers.
The comparison of simultaneous occurrences of oods in the study rivers (Fig. 8) revealed that such cases were common. Almost
half of them occurred in the Vistula and Danube (44 %), on the Vistula and the Elbe (40 %) and on Vistula and the Main (39 %). A
signicantly smaller correspondence in ood occurrences was noted between the Vistula and Norrstr¨
om rivers (only 10 %). The
simultaneous occurrence of oods on the Oder and other Central European and Swedish rivers revealed a slightly better correspon-
dence than did the results presented for the Vistula. The agreement in timing of occurrence of oods reached 50 %, 48 % and 44 %,
respectively, for the following pairs of rivers: Oder–Danube, Oder–Elbe and Oder–Main. This correspondence between oods occurring
on Oder River and rivers in Czech Republic (the Dyje, Morava and Vltava) is clearly smaller and ranges from 18 % (Dyje) to 29 %
(Morava).
In the last decade, two comprehensive studies have been published by Glaser et al. (2010) and Bl¨
oschl et al. (2020) summarizing
knowledge about ood occurrences in Europe in the past 500 years, and these warrant discussion here. They discovered different
numbers of ood-rich periods for Central Europe in 17–18th centuries, that only partly overlapped on other. Glaser et al. (2010)
identied two such periods (1640–1700, and 1730–1790), while Bl¨
oschl et al. (2020) found only one (1750–1800). Our results for
Poland (see Fig. 2a) reveal the existence of one ood-rich period (1650–1740), which is more consistent with the ndings presented by
Glaser et al. (2010). Also in most of the studied regions (Baltic Coast and Pomerania, Silesia, and Lesser Poland) we identied an
increase in oods frequencies in this time (Fig. 4). Only in Lesser Poland was there another ood-rich period noted in years 1775–1790
(Fig. 4f), which is consistent with the ood-rich periods identied by Glaser et al. (2010) and Bl¨
oschl et al. (2020). On the other hand, a
very good correspondence of results exists between the timings of ood occurrences within a year. Both Bl¨
oschl et al. (2020) and we in
this study have documented that oods in Central Europe dominate in summer and are least frequent in autumn (compare Fig. 5 herein
and Fig. 5 in Bl¨
oschl et al., 2020).
Due to their dire consequences, the assessment of extreme oods has been always considered of great importance for ood
occurrence studies. Thus, we evaluated the occurrences of extreme oods in this study with extreme oods in studies available for
Poland and some neighboring countries (Table S8). It is also important to mention that the criteria for classication of oods as
“extreme ood” in this study could differ from those utilized in other studies. In the Polish literature, a list of extreme oods has been
presented only by Kowalewski (2006). However, the number of extreme oods in that source is only seven, whereas, based on the
Br´
azdil et al. (2006b) classication (category 3), we documented as many as 69, 49 and 1 such cases in Vistula River, Oder River and
Baltic Coast rivers basins, respectively. Signicantly greater number of extreme oods were also distinguished in Poland (90, 73 and 4
oods in the basins of Vistula River, Oder River and Baltic Coast rivers, respectively) when the criterion proposed by Barriendos and
Coeur (2004) was used (category 2). The number of extreme oods in Germany was relatively high, with 85 oods in various river
basins, while in the other analyzed countries (Czech Republic, Sweden and Austria) the gures were markedly lower (20, 8 and 1,
respectively). The year with occurrences of extreme oods in Poland (this study) and in other countries are listed in Table S8.
In general, the analysis of spatial variability of ood occurrences showed that they were most numerous in the Silesia region of
Fig. 8. Number of oods in main Central Europe rivers and in Sweden in the period 1601–1800.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
10
south-western Poland (246, 36 %) and least numerous in the north-eastern Masuria-Podlasie region (6, less than 1 %). This discrep-
ancy is partly related to the availability of sources and historical weather notes, which are evidently richer in the Silesia region than in
the Masuria-Podlasie region. Furthermore, the Silesia region, particularly its Sudeten Mountains area, has a higher density of rivers
compared to lowland areas, making it more vulnerable to extreme events than other regions in Poland.
The comparison of ood occurrences in Poland during the 17th–18th centuries against such statistics available for the medieval
period (11th–15th centuries) (Ghazi et al., 2023a) and the 16th century (Ghazi et al., 2023b) revealed the existence of a rising trend. In
general, the number of oods in the 17th–18th centuries (335 oods in the 17th century and 343 in the 18th) was slightly higher than
in the 16th century (294 oods) and signicantly higher than in the 11th–15th centuries (166 oods). This is partly related to the
obvious increasing availability of historical sources and weather notes. The second possible reason is that, in the more distant past,
there existed a tendency to describe mainly oods deemed to be truly extreme. Our calculations of the intensity of oods in Poland in
the period 1001–1800 (this study and Ghazi et al., 2023a, b) based on the Br´
azdil et al. (2006b) classication support this opinion. We
found that most of the oods in the 17th–18th centuries were of the category “smaller, regional oods”, whereas in the 11th–15th and
16th centuries the category “above average or supra-regional oods” was more common. Another important nding is that, except in
the 16th century, oods were more common in the Vistula River basin than in the Oder River basin. The discrepancy, however, was
signicantly greater in medieval times than in the 17th–18th centuries. On the other hand, no changes throughout the time were noted
in the origins of oods. In all three studied subperiods, rain and its subtypes were the main causes of ood in Poland.
It is worth mentioning that, despite the fact that hydraulic and hydrological designers use measurement data such as ow rate in the
design of hydraulic structures, the ndings of this study, by providing information from the history of oods in the study area, can
provide useful information for the estimation of ood return periods.
5. Conclusions
The presented broad analysis of historical oods in Poland during the 17th–18th centuries based on documentary evidence im-
proves our understanding of historical oods in Poland and Central Europe. To achieve this objective, a comprehensive database was
constructed that included an indexing of each ood’s intensity and origin. It is worth noting that this study completed the knowledge
gap for historical oods in Poland before the instrumental periods and will play a crucial role in bridging the knowledge gap for the
assessment of historical oods in Europe before the 19th century. We would like to emphasize that in this study we tried to provide all
available information for historical oods in Poland in the 17th–18th centuries from existing sources, and the ndings provided in
these results are based on all documentary evidence that has been made available to date. Thus, it is important to remember that the
results for the frequency of oods may be modied and updated in the future if new sources come to light.
The primary ndings and summary of this study are outlined as below:
1. Overall, 678 oods occurred in Poland in the 17th–18th centuries.
2. The frequency of oods in the Vistula River basin (51 %) is slightly higher than
in the Oder River basin (47 %).
3. Floods occurred most in Silesia (246, 36 %), Baltic Coast and Pomerania (24 %) and Lesser Poland (21 %).
4. Floods mostly belong to the “smaller regional oods” category based on the Br´
azdil et al. (2006b) classication and the
“extraordinary” category in the Barriendos and Coeur (2004) classication.
5. The greatest number of oods by season occurred in summer, of which, especially in July.
6. Rain and its subtypes were the dominant reason for ood occurrences.
7. Overall, in Poland, 119 “extreme oods” occurred based on the Br´
azdil et al. (2006b) classication (category 3) and 167 oods
classied as “extreme oods” in the Barriendos and Coeur (2004) classication (category 2).
8. Comparison of the spatial coherency of oods in Poland with other Central Europe countries showed that there is a good coherency
with oods occurring in Germany and Czech Republic.
Financial support
This research was funded by a grant from the National Science Centre, Poland, project (no: 2020/37/B/ST10/00710). The funding
for publication costs (open-access) was provided by the "Excellence Initiative - Research University" program (IDUB) at the Nicolaus
Copernicus University in Toru´
n, Poland.
CRediT authorship contribution statement
Babak Ghazi: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Software, Visualization, Writing –
original draft, Writing – review & editing. Rajmund Przybylak: Conceptualization, Methodology, Investigation, Data curation,
Formal analysis, Funding acquisition, Project administration, Validation, Supervision, Writing – original draft, Writing – review &
editing. Piotr Oli´
nski: Investigation, Data curation, Validation. Michał Targowski: Data curation, Validation. Janusz Filipiak: Data
curation, Validation. Aleksandra Pospieszy´
nska: Data curation, Validation, Software, Visualization.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
11
Declaration of Competing Interest
The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to
inuence the work reported in this paper.
Data Availability
The dataset of ood records for Poland in the 17th–18th centuries is available at (https://doi.org/10.18150/VLPAFG). Other data
used in this research are available from the corresponding author, upon reasonable request.
Acknowledgments
The authors would like to acknowledge the National Science Centre, Poland, for providing a grant project (no: 2020/37/B/ST10/
00710) for this project and "Excellence Initiative - Research University" program (IDUB) at the Nicolaus Copernicus University in
Toru´
n for providing funding to publish this research as an open-access paper. We would like to thank Ms. Katarzyna Bogda´
nska for her
assistance in collecting data for this research.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ejrh.2024.101796.
References
Barriendos, M., Coeur, D., 2004. Flood data reconstruction in historical times from non-instrumental sources in Spain and France. Systematic, Palaeoood and
Historical Data for the Improvement of Flood Risk Estimation. Methodological Guidelines, edited by: Benito, G. and Thorndycraft, VR, Centro de Ciencias
Medioambientales, Madrid, Spain, pp. 29–42.
Benito, G., Castillo, O., Ballesteros-C´
anovas, J.A., Machado, M., Barriendos, M., 2021. Enhanced ood hazard assessment beyond decadal climate cycles based on
centennial historical data (Duero basin, Spain). Hydrol. Earth Syst. Sci. 25 (12), 6107–6132.
Bhat, M.S., Ahmad, B., Alam, A., Farooq, H., Ahmad, S., 2019. Flood hazard assessment of the Kashmir Valley using historical hydrology. J. Flood Risk Manag. 12,
e12521.
Biela´
nski, A.K. (1997). Materiały do historii powodzi w dorzeczu G´
ornej Wisły: na podstawie rękopisu Adama Kazimierza Biela ´
nskiego i materiał´
ow Jana Fiszera.
Bl¨
oschl, G., Hall, J., Parajka, J., Perdig˜
ao, R.A., Merz, B., Arheimer, B., Borga, M., 2017. Changing climate shifts timing of European oods. Science 357 (6351),
588–590.
Bl¨
oschl, G., Hall, J., Viglione, A., Perdig˜
ao, R.A., Parajka, J., Merz, B., Bilibashi, A., 2019. Changing climate both increases and decreases European river oods.
Nature 573 (7772), 108–111.
Bl¨
oschl, G., Kiss, A., Viglione, A., Barriendos, M., B¨
ohm, O., Br´
azdil, R., Macdonald, N., 2020. Current European ood-rich period exceptional compared with past 500
years. Nature 583 (7817), 560–566.
Br´
azdil, R., Dobrovolný, P., Elleder, L., Kakos, V., Kotyza, O., Kvˇ
etoˇ
n, V., Tolasz, R., 2005b. Historical and recent oods in the Czech Republic. Masaryk Univ. Czech
Hydrometeorol. Inst.
Br´
azdil, R., Dobrovolný, P., Elleder, L., Kakos, V., Kotyza, O., Kvˇ
etoˇ
n, V., … Tolasz, R. (2005a). Historick´
e a souˇ
casn´
e povodnˇ
e v ˇ
Cesk´
e republice: Masarykova
univerzita v Brnˇ
e, ˇ
Ceský hydrometeorologický ústav v Praze.
Br´
azdil, R., Dobrovolný, P., Elleder, L., Kakos, V., Kotyza, O., Kvĕtoˇ
n, V., M´
ackova, J., Müller, M., ˇ
Stekl, J., Tolasz, R., H, a V., 2006a. Historical and recent oods in
the Czech Republic: causes, seasonality, trends, impacts. Flood risk management: Hazards, Vulnerability and Mitigation Measures. Springer, pp. 247–259.
Br´
azdil, R., & Kirchner, K. (2007). Vybran´
e pˇ
rírodní extr´
emy a jejich dopady na Moravˇ
e a ve Slezsku (Selected Natural Extremes and Their Impacts in Moravia and
Silesia). Masarykova univerzita Brno.
Br´
azdil, R., Kundzewicz, Z.W., Benito, G., 2006b. Historical hydrology for studying ood risk in Europe. Hydrol. Sci. J. 51 (5), 739–764.
Br´
azdil, R., Pster, C., Wanner, H., Storch, H.V., Luterbacher, J., 2005c. Historical climatology in Europe–the state of the art. Clim. Change 70, 363–430.
Br´
azdil, R., ˇ
Rezníˇ
ckov´
a, L., Val´
aˇ
sek, H., Havliˇ
cek, M., Dobrovolný, P., Soukalov´
a, E., Skokanova, H., 2011. Fluctuations of oods of the River Morava (Czech Republic)
in the 1691–2009 period: interactions of natural and anthropogenic factors. Hydrol. Sci. J.–J. Des. Sci. Hydrol. 56 (3), 468–485.
Brunner, M.I., Melsen, L.A., Newman, A.J., Wood, A.W., Clark, M.P., 2020. Future streamow regime changes in the United States: assessment using functional
classication. Hydrol. Earth Syst. Sci. 24 (8), 3951–3966.
Cœur, D., Lang, M., 2008. Use of documentary sources on past ood events for ood risk management and land planning. Comptes Rendus Geosci. 340 (9-10),
644–650.
Cyberski, J., Grze´
s, M., Gutry-Korycka, M., Nachlik, E., Kundzewicz, Z.W., 2006. History of oods on the River Vistula. Hydrol. Sci. J. 51 (5), 799–817.
Embleton, C., Embleton-Hamann, C., 1997. Geomorphological hazards of Europe. Elsevier.
Fal, B., Dąbrowski, P., 2001. Dwie´
scie lat obserwacji i pomiar´
ow hydrologicznych Wisły w Warszawie. Cz II. Przepływy Wisły w Warszawie. Gospod. Wodna 503–510.
García-Barr´
on, L., Aguilar-Alba, M., Morales, J., Sousa, A., 2023. Classication of the ood severity of the Guadalquivir River in the Southwest of the Iberian Peninsula
during the 13th to 19th centuries. Atm. ´
oSfera 36 (1), 1–21.
Ghazi, B., Przybylak, R., Oli´
nski, P., Bogda´
nska, K., Pospieszy´
nska, A., 2023a. The frequency, intensity, and origin of oods in Poland in the 11th–15th centuries based
on documentary evidence. J. Hydrol., 129778
Ghazi, B., Przybylak, R., Oli´
nski, P., Chorą˙
zyczewski, W., Pospieszy´
nska, A., 2023b. An assessment of ood occurrences in Poland in the 16th century. J. Hydrol. Reg.
Stud. 50, 101597.
Glaser, R., Riemann, D., Sch¨
onbein, J., Barriendos, M., Br´
azdil, R., Bertolin, C., van Engelen, A., 2010. The variability of European oods since AD 1500. Clim. Change
101 (1), 235–256.
Glaser, R., Stangl, H., 2003. Historical oods in the Dutch Rhine delta. Nat. Hazards Earth Syst. Sci. 3 (6), 605–613.
Grze´
s, M. (1991). Zatory i powodzie zatorowe na dolnej Wi´
sle: mechanizmy i warunki=Ice jams and oods on the lower Vistula river: mechanism and processes.
Grze´
s, M. (2008). Historia powodzi na Wi´
sle w ´
swietle tablic wielkich w ´
od. http://www.wielkawoda.umk.pl/.
Hohensinner, S. (2015). Historische Hochw¨
asser der Wiener Donau und ihrer Zubringer: Zentrum für Umweltgeschichte.
Hundecha, Y., Parajka, J., Viglione, A., 2020. Assessment of past ood changes across Europe based on ood-generating processes. Hydrol. Sci. J. 65 (11), 1830–1847.
B. Ghazi et al.

Journal of Hydrology: Regional Studies 53 (2024) 101796
12
Kasprzak, M., 2010. Wezbrania i powodzie na rzekach Dolnego Slaska. Wyjatkowe Zdarzenia Przyr. Na Dolnym Slasku I Ich Skutki, Ed. : Migon, P., Rozpr. Nauk. Inst.
Geogr. I Rozw. Reg. Uniw. Wroc. łAwskiego 14, 81–140.
Kjeldsen, T., Macdonald, N., Lang, M., Mediero, L., Albuquerque, T., Bogdanowicz, E., Fleig, A., 2014. Documentary evidence of past oods in Europe and their utility
in ood frequency estimation. J. Hydrol. 517, 963–973.
Kowalewski, Z., 2006. Powodzie w Polsce-rodzaje, występowanie oraz system ochrony przed ich skutkami. Woda-´
srodowisko-obsz. Wiej. 6, 207–220.
Lambor, J., 1954. Klasykacja typ´
ow powodzi i ich przewidywanie. Gospod. Wodna 14 (4), 129–131.
Majewski, A., 1993. Kronika Powodzi w delcie Wisły,[w:] Uwarunkowania przyrodnicze i społeczno-ekonomiczne zagospodarowania dolnej Wisły, red. Z. Churski
Instytut Geograi, Uniwersytet Mikołaja Kopernika, Toru´
n.
Merz, B., Bl¨
oschl, G., Vorogushyn, S., Dottori, F., Aerts, J.C., Bates, P., Lall, U., 2021. Causes, impacts and patterns of disastrous river oods. Nat. Rev. Earth Environ. 2
(9), 592–609.
Mudelsee, M., B¨
orngen, M., Tetzlaff, G., Grünewald, U., 2003. No upward trends in the occurrence of extreme oods in central Europe. Nature 425 (6954)), 166–169.
Mudelsee, M., Deutsch, M., B¨
orngen, M., Tetzlaff, G., 2006. Trends in ood risk of the River Werra (Germany) over the past 500 years/Tendances du risque
d
′
inondation dans la vall´
ee de la rivi`
ere Werra (Allemagne) durant les 500 derni`
eres ann´
ees. Hydrol. Sci. J. 51 (5), 818–833.
Nachlik, E., Kundzewicz, Z.W., 2016. History of oods on the Upper Vistula. Flood Risk Up. Vist. Basin 279–292.
Pster, C. (2002). Am Tag danach: zur Bew¨
altigung von Naturkatastrophen in der Schweiz 1500-2000. (No Title).
Rets¨
o, D., 2015. Documentary evidence of historical oods and extreme rainfall events in Sweden 1400–1800. Hydrol. Earth Syst. Sci. 19 (3), 1307–1323.
Rohde, M.M., 2023. Floods and droughts are intensifying globally. Nat. Water 1 (3), 226–227.
Rohr, C., 2006. Measuring the frequency and intensity of oods of the Traun River (Upper Austria), 1441–1574. Hydrol. Sci. J. 51 (5), 834–847.
Rohr, C., 2007. Extreme Naturereignisse im Ostalpenraum: Naturerfahrung im Sp¨
atmittelalter und am Beginn der Neuzeit. Rev. De. L
′
ifha, Date De. Par. De. L
′
œuvre.
Rottler, E., Bronstert, A., Bürger, G., Rakovec, O., 2023. Rhine ood stories: Spatio-temporal analysis of historic and projected ood genesis in the Rhine River basin.
Hydrol. Process. 37 (6), e14918.
Sturm, K., Glaser, R., Jacobeit, J., Deutsch, M., Brazdil, R., Pster, C., Wanner, H., 2001. Hochwasser Mitteleur. seit 1500 und ihre Bezieh. Zur. atmosph¨
arischen
Zirkulation.
Wilhelm, B., Ballesteros C´
anovas, J., Macdonald, N., Toonen, W., Baker, V., Barriendos, M., Denniston, R., 2019. Interpreting historical, botanical, and geological
evidence to aid preparations for future oods. WIREs Water 6, e1318.
B. Ghazi et al.