Tree-ring reconstruction of the level of endorheic lakes in the south of West Siberia

Abstract

The paper presents the dendrochronological method applied in reconstructing the changes in the water level of endorheic lakes from arid areas (on the example of two large endorheic lakes Kulundinskoye and Chany). Chronologies of annual ring width of Scots pine (Pinus sylvestris L.), growing in the pine belt forests in the south of West Siberia, were used as indicators of climate change. To obtain the longest reconstructions, we used the lengthen chronologies. For analysis, specimens of living trees and old wooden houses in the Altai Krai villages (adjacent to pine-belt forests) were used. Due to the dendrohydrological analysis, we have determined the relationship between the lake level and the radial pine growth. We have revealed a one-two-year lag in the response of hydrological objects to climate fluctuations. Correlation between the tree-ring chronologies and the average annual water level of Lake Kulundinskoye occurred more often due to proximity of both geosystems, i.e. the lake and pine-belt forests. At the same time, there was the best correlation for Lake Chany and the drysteppe chronology Uglovskoye 42_LL. In the course of investigations, we have extended the series for lakes Chany and Kulundinskoye up to 1584 and 1763, respectively. Spectral analysis of the series of the water level in both lakes allowed us to identify common cycles (i.e. 25-26, 40-46 and 80-100-year ones). For Lake Kulundinskoye, the 8 and 11-year cycles turned out to be the most significant. The analysis of secular cycles of hydrological series enabled to specify the trends in water-level changes of the studied lakes in the future.

Full text

IOP Conference Series: Earth and Environmental Science
PAPER • OPEN ACCESS
Tree-ring reconstruction of the level of endorheic lakes in the south of
West Siberia
To cite this article: N V Rygalova and V P Galakhov 2019 IOP Conf. Ser.: Earth Environ. Sci. 381 012081
View the article online for updates and enhancements.
This content was downloaded from IP address 45.11.233.253 on 23/11/2019 at 12:34

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
Geographical Foundations
IOP Conf. Series: Earth and Environmental Science 381 (2019) 012081
IOP Publishing
doi:10.1088/1755-1315/381/1/012081
1
Tree-ring reconstruction of the level of endorheic lakes in the
south of West Siberia
N V Rygalova1 and V P Galakhov2
1Altai State University, Russia, Barnaul
2Institute for Water and Environmental Problems SB RAS, Russia, Barnaul
E-mail: natalia.ml@mail.ru
Abstract. The paper presents the dendrochronological method applied in reconstructing the
changes in the water level of endorheic lakes from arid areas (on the example of two large
endorheic lakes Kulundinskoye and Chany). Chronologies of annual ring width of Scots pine
(Pinus sylvestris L.), growing in the pine belt forests in the south of West Siberia, were used as
indicators of climate change. To obtain the longest reconstructions, we used the lengthen
chronologies. For analysis, specimens of living trees and old wooden houses in the Altai Krai
villages (adjacent to pine-belt forests) were used. Due to the dendrohydrological analysis, we
have determined the relationship between the lake level and the radial pine growth. We have
revealed a one-two-year lag in the response of hydrological objects to climate fluctuations.
Correlation between the tree-ring chronologies and the average annual water level of Lake
Kulundinskoye occurred more often due to proximity of both geosystems, i.e. the lake and
pine-belt forests. At the same time, there was the best correlation for Lake Chany and the dry-
steppe chronology Uglovskoye 42_LL. In the course of investigations, we have extended the
series for lakes Chany and Kulundinskoye up to 1584 and 1763, respectively. Spectral analysis
of the series of the water level in both lakes allowed us to identify common cycles (i.e. 25-26,
40-46 and 80-100-year ones). For Lake Kulundinskoye, the 8 and 11-year cycles turned out to
be the most significant. The analysis of secular cycles of hydrological series enabled to specify
the trends in water-level changes of the studied lakes in the future.
1. Introduction
A water level of endorheic lakes, which depends on moistening of the territory, is among the
indicators of regional climate change [1-4]. To understand the dynamics of level regime variability in
endorheic lakes, the analysis of long hydrological series is required. The main source of such data is
the in-situ observations implemented at gauging stations. However, their duration often hampers the
identification of long-term (e.g. secular, etc.) cyclical fluctuations. Due to the common limiting factor
(i.e. territory moistening), dendrochronological data may contribute to solving this problem. In arid
regions, the increase in precipitation amount favors the radial growth of trees and the increase in the
water level of a lake. On the contrary, the increase in summer temperature leads to a decrease in the
annual ring width and the water level due to intensive water evaporation from the reservoir surface.
Thus, being climate-related, the relationships between these features do certainly exist.
A large body of dendrohydrological research carried out in different regions [5-10] is evidence of
the method reliability, therefore, and it can be recommended for applying under insufficient
moistening conditions when a climatic signal in geosystems is more pronounced.

Geographical Foundations
IOP Conf. Series: Earth and Environmental Science 381 (2019) 012081
IOP Publishing
doi:10.1088/1755-1315/381/1/012081
2
2. Models and methods
In this paper presents the dendrohydrological analysis of the dynamics of the radial growth (annual
ring width) of Scots pine (Pinus sylvestris L.) from the pine-belt forests in the forest-steppe and steppe
(arid and dry) areas of Altai Krai. The island nature of pine-belt forests, growing outside the Scots
pine dense area, makes the trees sensitive indicators, which are used for the reconstruction of past
climatic conditions of the vegetation period and climate-driven processes, e.g. dynamics of parameters
of hydrological objects.
In the south of West Siberia, the area of internal drainage with a large number of lakes was formed.
For instance, these are the largest lakes Kulundinskoye (Altai Krai) and Chany (Novosibirsk Oblast),
whose water level, along with the 30 tree-ring width (TRW) chronologies for the pine-belt forests
[11], have become the basis of our studies. The chronologies were obtained through the generally
accepted method [12]. The extended chronologies were used to reconstruct the dynamics of
hydrological indicators. For this purpose, the specimens of living trees from the pine-belt forests, as
well as wood from historic wooden houses in the Altai villages, were collected.
Salty Lake Kulundinskoye is located almost in the center of the Kulunda lowland, in the west of
Altai Krai (in the steppe zone). The total area of the lake is approximately 728 km2; its average depth –
3.2 m. A series of field observations for 1934-1969 (except for 1959-1960) implemented at the
gauging station Belgrad is available. Lake Chany is situated in the Baraba plain, specifically in its
forest-steppe zone. The reservoir consists of several independent limnosystems (stretches). The size of
the water area is currently 1500 km2. The depth ranges from 1.4-1.9 m; in the southeast of brackish
Lake Yarkul it reaches 4.8-8.5 m [13]. The 1934-2008 observation period (data from the gauging
station Kvashnino) was analyzed. To ensure the accurate comparison of the results for both lakes
during the correlation analysis, we selected the period between 1934 and 1969.
For establishing relationships in the functioning of different-scale geosystems, the asynchronous
response to external factors should be properly taken into account. Standardized tree-ring series
ARSTAN (which, unlike residual chronologies, preserve long-term cycles) were used to provide the
maximum similarity in hydrological/biological objects response to climate changes. It is worth noting
that lake level fluctuations are slower in comparison with tree growth dynamics. Along with the
recorded increase in the average correlation, dendrohydrological relationships are traced over the next
two years (the delay in climate response of the lake level) (table 1). During the third year, the
relationships grow weak. Their maximum is typically observed with a one-two year delay relative to
changes in the TRW (absolute maximum occurs in one year). Compared to Lake Chany, for the water
level of Lake Kulundinskoye, significant correlations with chronologies prevailed, and the average
correlation was higher. This can be explained by the territorial proximity of two geosystems (Lake
Kulundinskoye and the adjacent pine-belt forest). In both cases, there are no traces of spatial
coherence with some TRW chronologies.
Table 1. A generalized description of the dendrohydrological relationship of chronologies of pine
belt forests and water level series for lakes Chany and Kulundinskoye (1934-1969).
The shift of
hydrological series
relative to TRW
chronologies
Lake Kulundinskoye
Lake Chany
coefficient of average correlation of water level with all chronologies;the
maximum correlation between water level and TRW is given in parentheses
Synchronous
0.20 (0.56 with Uglovskoye 42_LL)
0.06 (0.67 with Uglovskoye42_LL)
A year ago
0.33 (0.67 with Mamontovo_Buk)
0.16 (0.79 with Uglovskoye 42_LL)
2 years ago
0.36 (0.66 with Mamontovo _Buk)
0.20 (0.77 with Uglovskoye 42_LL)
For Lake Chany, the best correlation was obtained for the dry-steppe chronology Uglovskoye
42_LL (with a one-year shift in hydrological series). This TRW chronology, which was built with
living pine trees from the southern part of the Barnaul belt forest (1997-1844), was extended using

Geographical Foundations
IOP Conf. Series: Earth and Environmental Science 381 (2019) 012081
IOP Publishing
doi:10.1088/1755-1315/381/1/012081
3
historic wood sampled from three old wooden houses in the Laptev Log village (Altai Krai) adjacent
to the model site of the pine-belt forest. The total chronology length of 415 years was extended (up to
1584). The standard deviation of the extended generalized chronology was 0.3, which is indicative of
its climatic sensitivity.
For Lake Kulundinskoye, TRW chronology Mamontovo_Buk showed the best correlation (also
with a one-year shift in hydrological series). The chronology was built with living trees that grow in
the middle part of the Kasmala belt forest (1852-2002) as well as logs from the house that was built in
the early 20th century in the Bukanskoye village adjacent to the sampling site. It allowed us to extend
the series up to the year 1762.
The regression equation for the water level of Lake Kulundinskoye and TRW chronology
Mamontovo for the 1934-1969 period is as follows:
02.11472.110 += yx
,
where x is the average annual water level of the lake, y − the index of the annual increment. The
correlation coefficient is 0.67, the coefficient of determination is 0.44.
The regression equation for the water level of Lake Chany and TRW chronology Uglovskoye
42_LL was derived for the same period (note: the data available up to 2008 provide visual verification
of the developed model) and has the following form:
33.1612.141 += yx
,
where x is the average annual water level of the lake, y − the index of the annual increment. The
correlation coefficient is 0.79, the determination one is 0.63.
The reconstructed series for the years 1934-2003 generally coincide with field observations.
However, in the 1970s, a discrepancy between the reconstructed and the natural series was recorded,
and the total correlation for both series in 1934-2003 was 0.37 despite their great visual similarity. In
1971-1972, the dams separating the Yudinsky stretch from the main part of the lake were constructed
[14]. Perhaps this hydrotechnical intervention influenced the synchronism in the variability of two
natural objects.
3. Results and discussion
During the studies, we have reconstructed a series of the water level for lakes Chany and
Kulundinskoye up to the years 1584 and 1762, respectively. The reconstructed series have a
pronounced cyclicism of water level fluctuations. For a series of Lake Kulundinskoye, the 8 and 11-
year cycles are the most significant. The 17-, 26.5-, 40-, and 80-year cycles are also essential. The 8-
and 11-year cycles occur in dynamics, whereas an 80-year cycle stands out when smoothing the 30-
year moving average (Figure 1). At present, the 80-year cycle descends, though in the mid-20s and
early 30s of the 21st century it is expected to be ascending.
Figure 1. Reconstructed and actual series of the average annual water level of
Lake Kulundinskoye

Geographical Foundations
IOP Conf. Series: Earth and Environmental Science 381 (2019) 012081
IOP Publishing
doi:10.1088/1755-1315/381/1/012081
4
The reconstructed series for Lake Chany also show cycles of different duration (figure 2). The 12-,
21-26-, 32-37-, 41-46-, 52-, and 100-year cycles (specifically the 103-year cycle) in water level
fluctuations prevail. The reconstructed series covers the period of the Little Ice Age, during which
there were positive and negative fluctuations in the water level of the lake. The specificity of the
processing data method (standardization of raw tree-ring chronologies) is that it results in removing
long-term fluctuations from a series. For this reason, only secular and intra-secular cycles are present
in the reconstructed hydrological series.
Figure 2. Reconstructed and actual series of the average annual water level in Lake
Chany
To obtain more reliable results, it is necessary to extend the indication basis (TRW chronologies)
into the past and increase the replication of individual series. Nowadays, proper efforts are underway.
4. Conclusion
This dendrohydrological study suggests the reconstructed changes in water level fluctuations of large
closed-basin lakes in West Siberia (i.e. lakes Chany and Kulundinskoye until the late 16th century and
mid-18th century, respectively). The reconstructed series show cyclicity (11-, 25-26, 40-46, and 80-
100-year cycles) similar to that of natural-climatic rhythms driven by solar activity. Overall, the
dendrochronological method application is quite promising for studying arid and poorly moistened
areas. Moreover, the formation of the TRW base for the forest-steppe and steppe zones could ensure
reconstructing the climatic and climate-related processes. Extension of chronologies using historic
wood is of considerable importance in providing the longer reconstruction of hydrological events of
the past.
Acknowledgments
The paper is based on the data analysis made within the framework of RFBR project No. 18-05-
00694_а.
References
[1] Agupov A 1958 Lakes as indicators of changes in moistening of river basin territories Bulletin
of the USSR Academy of Sciences: Seria Geography 1 48–60

Geographical Foundations
IOP Conf. Series: Earth and Environmental Science 381 (2019) 012081
IOP Publishing
doi:10.1088/1755-1315/381/1/012081
5
[2] Galakhov V 2012 Evaluation of territory moistening in the south of West Siberia (by water level
fluctuations in Lake Chany) Bulletin of the RGS 1(144) 59–63
[3] Nekhaichik V 1965 Impact of fluctuations in the total moistening of the Kulunda steppe on
water balance of lake Kulundinskoye Izvestiya of RGO 1(97) 75–7
[4] Shnitnikov A 1968 Interdecadal Variability of Total Moistening Components (Leningrad:
Nauka) p 264
[5] Vakhnina I, Golyatina A and Noskova E 2018 Indicators of climate change in the steppe zone of
South-Eastern Transbaikalia Proc. of the Int. Scientific-Practical Conf. and Symposium
Dedicated to the 100th Anniversary of Nature Reserves and the Year of Ecology in Russia:
The Silk Road. The Trans-Sib. Integration Routes: Economy, Ecology (Chita: Institute of
Natural Resources, Environment and Cryology SB RAS) pp 34–7
[6] Levi K, Miroshnichenko A, Kozyreva E, Voronin V and Zuev F 2016 Dynamics of lake’s level
regime Proc. of III Russian Meeting and II Russian Youth School on Modern Geodynamics:
Modern Geodynamics of Central Asia and Extreme Natural Events: Results of Quantitative
Research (Irkutsk: Institute of the Earth’s Crust SB RAS) pp 267–9
[7] Pakal’nis R 1972 Application of dendroclimatology methods in determining water level
fluctuations in lakes of Eastern Lithuania Proc. of II USSR Meeting on Dendrochronology
and Dendroclimatology: Dendroclimatochronology and Radiocarbon (Kaunas: Lithuanian
Institute of Botany) pp 198–204
[8] George S 2002 Hydroclimatic Change in Southern Manitoba Since A.D. 1409 Inferred from
Tree Rings Quaternary Research 58 103–11 doi:10.1006/qres.2002.2343
[9] Meko D 2006 Tree ring inferences on water-level fluctuations of Lake Athabasca Canadian
Water Resources Journal 4(1) 1–20
[10] Stockton C and Fritts H 1973 Long-term Reconstruction of Water Levels Changes for Lake
Athabasca by Analysis of Tree Rings JAWRA Journal of the American Water Resources
Association 9 1006–27
[11] Malysheva (Rygalova) N and Bykov N 2011 Dendrochronological Studies of Pine Forests in
the South of West Siberia (Barnaul: Azbuka) p 125
[12] Shiyatov S et al 2000 Methods of Dendrochronology Part I (Krasnoyarsk: publishing house
KrasSU) p 80
[13] Vasiliev O, Kazantsev V, Popov P and Kirillov V 2005 General nature characteristic and
environmental problems of Chany and Kulundinskoye lake systems and their basins Siberian
Ecological Journal 2 167–73
[14] Vasiliev O, Savkin V and Saprykina Ya 2006 Analysis of level fluctuations in lake Chany RAS
Reports 4(407) 533–6