Next Article in Journal
Understanding the Determinants of Lane Inefficiency at Fully Actuated Intersections: An Empirical Analysis
Next Article in Special Issue
Using Remote Sensing to Assess the Vegetation Cover of a Protected Salt Marsh Subjected to Artificial Recharge and Groundwater Abstractions during the Period 1925–2022 (Alicante, SE Spain)
Previous Article in Journal
Spatial Distribution and Accessibility Analysis of Primary School Facilities in Mega Cities: A Case Study of Chengdu
Previous Article in Special Issue
Preliminary Study of the Occurrence of Microplastics in the Sediments of the Rzeszów Reservoir Using the Laser Direct Infrared (LDIR) Method
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 2
10.3390/su16020725
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Warming Climate-Induced Changes in Lithuanian River Ice Phenology

by
Diana Šarauskienė
,
Darius Jakimavičius
,
Aldona Jurgelėnaitė
and
Jūratė Kriaučiūnienė
*
Laboratory of Hydrology, Lithuanian Energy Institute, Breslaujos St. 3, LT-44403 Kaunas, Lithuania
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(2), 725; https://doi.org/10.3390/su16020725
Submission received: 1 December 2023 / Revised: 11 January 2024 / Accepted: 12 January 2024 / Published: 14 January 2024
(This article belongs to the Special Issue Sustainability in Water Resources, Water Quality, and Architecture)
Browse Figures
Versions Notes

Abstract

:
Due to rising surface air temperatures, river ice is shrinking dramatically in the Northern Hemisphere. Ice cover during the cold season causes fundamental changes in river ecosystems and has important implications for nearby communities and industries. Changes caused by climate warming, therefore, affect the sustainability of key resources, livelihoods, and traditional practices. Thus far, too little attention has been paid to research into the phenomenon of river ice in the Baltic States. Since the observational data of the last sixty years are currently available, we took advantage of the unique opportunity to assess ice regime changes in the gauged rivers by comparing two climatological standard normals. By applying statistical methods (Mann–Kendall, Pettitt, SNHT, Buishand, von Neumann, and Wilcoxon rank sum tests), this study determined drastic changes in ice phenology parameters (freeze-up date, ice break-up date, and ice cover duration) of Lithuanian rivers in the last thirty-year period. The dependence of the selected parameters on local climatic factors and large-scale atmospheric circulation patterns was identified. It was established that the sum of negative air temperatures, as well as the North Atlantic Oscillation, East Atlantic, and Arctic Oscillation indices, have the greatest influence on the ice regime of Lithuanian rivers.

1. Introduction

Global climate change is one of the most severe challenges facing modern society. Climate warming, an increasing number of extreme weather events, melting glaciers, rising sea levels, drastic changes in wildlife habitats and populations, and many other effects of climate change are leading our planet to unprecedented chaos step by step. The increasing temperature of the Earth is widely acknowledged to be due to elevated carbon dioxide levels in the atmosphere. Every year, there is growing evidence that CO2 emission by humans is the main factor in the rise of the Earth’s temperature. Although it cannot be fully proven, it is suggested that the relationship between CO2 concentration and the planet’s temperature is as strong as ever [1]. In 2017, global warming reached approximately 1 °C above pre-industrial levels (i.e., relative to 1850–1900), increasing by 0.2 °C per decade [2]. The period from 2015 to 2022 is likely to be the warmest eight-year period on record [3]. The air temperature rises every season; however, in the Northern Hemisphere (NH), the signs of warming in the cold season are often clearly expressed by the significantly shorter duration of ice cover on water bodies or the absence of it. In this part of the planet, ice has been found to seasonally cover 58% of the total length of rivers [4]. The ice-covered surface of the water during the cold season causes fundamental changes in river ecosystems. It has important consequences for communities and industries closely associated with these frozen waterways. From an engineering perspective, river ice events such as ice-jam floods pose the risk of infrastructure being damaged or destroyed [5]. Research shows that river ice is drastically declining globally and will continue to decline due to increasing surface air temperatures [6]. Since a vast number of scientific studies [7,8,9,10,11] indicate that alterations in river ice processes follow air temperature trends, the ice phenomenon can be used to indicate ongoing climate change.
Due to the strong climate/river ice relationship, pronounced variations in ice cover and seasonality exist. Because this over-time-established recurrence of ice processes is being disturbed, the timing of freeze-up and break-up in rivers and lakes (known as ice phenology) is often used to estimate the rate of ice loss. Using ice data of 39 lakes and rivers around the Northern Hemisphere available for 150 years (from 1846 to 1995), the authors of [12] found that changes in freeze-up dates averaged 5.8 days per 100 years later, and changes in break-up dates averaged 6.5 days per 100 years earlier. A recent study [13] of 3510 ice time series across 678 NH lakes and rivers estimated an increase of 0.63 days per decade in annual open-water days between 1931 and 2005. Using satellite observations, the authors of [6] estimated that the mean duration of river ice cover will decrease by 6.10 ± 0.08 days for each 1 °C increase in the global mean surface air temperature. The results of available regional or rare country-wide investigations follow similar trends: in Mongolia [14], the Asian part of Russia [15], Sweden [16], and Latvia [17], an estimated shift in river ice freeze-up and break-up dates is measured in several days or even months.
Many scientific publications are based on observations of individual rivers (or more frequently, separate reaches). These selected rivers usually have long observational data series that can reveal long-term changes in the ice regime. Data on the Lower Vistula section (Poland) starting from 1860 indicate a pronounced reduction in the duration of ice cover: the number of days with permanent ice cover decreased from about 70 to almost zero [18]. In the lower part of the Danube River, ice observations spanning 1837–2016 show a significant trend of an average 28-day decrease per century and more frequent ice-free winters in recent decades [9]. A unique record of ice break-up dates for the Torne River (Finland) goes back centuries [19]. Available data from 1693 to 2013 revealed a higher rate of earlier break-up: −0.30 days per decade before the breakpoint around 1866 to −0.66 days per decade after the break-point in this river. A study of ice regime dynamics in the lower reaches of the Nemunas River (Lithuania), based on ice observations from 1812 to 2000, identified significant negative break-up and positive freeze-up trends since the first half of the 19th century [20].
An increasing number of studies testify to the growing scale of climate change effects on the freshwater ice regime. The IPCC Special Report on the Ocean and Cryosphere in a Changing Climate [21], based on thousands of scientific publications, dispels the last doubts, if any, about the effects of a changing climate on cryosphere components.
Thus far, too little attention has been paid to research the river ice phenomenon in the Baltic States. The Second Assessment of Climate Change for the Baltic Sea Basin [22] briefly concluded that river ice duration in the Baltic Sea basin is decreasing due to later onset dates and earlier break-up dates. The literature search revealed several recent studies of the ice regime of Latvian and Lithuanian rivers [17,20,23]; however, they did not estimate changes in the river ice regime over the past decades and focused mainly on large rivers, leaving smaller rivers underestimated. The present study is partly a continuation of the previous study [24]; it includes a larger sample of more diverse rivers and extended data sets. Data covering additional years are currently available; therefore, it is a unique opportunity to assess changes in the river ice regime over the past sixty years, using the most recent 30-year period (1991–2020) and the period of 1961–1990 recommended as a standard reference for long-term climate change assessments [25]. This study aims to evaluate the changes in the phenological parameters of Lithuanian river ice by comparing the data of two periods of climatological standard normal and analyzing their dependence on air temperature and teleconnection patterns.

2. Materials and Methods

2.1. Study Area and Data

Lithuania is located in the western part of the Eastern European Plain. The country is in the northern part of the temperate climate zone. The climate is transitional between mild Western European and continental Eastern European. Two factors determine the country’s climate: the advection of air masses from the west and the warming effect of the Baltic Sea. In the cold season, the predominant air mass is from the west; however, an advection of air masses from the south often occurs.
Significant climatic differences are defined in the territory of Lithuania. The western part of the country features the influence of the Atlantic climate with high precipitation, whereas the eastern part is affected by a continental climate with lower precipitation and typically more severe winters. Due to the influence of the Atlantic Ocean and local relief, in winter, the air temperature has a meridional distribution; in the eastern part of Lithuania, the surface air temperature is lower than in the western [26]. Meanwhile, in summer, the temperature difference is insignificant. In Lithuania, the mean annual temperature for 1961–2020 was 6.8 °C, while in the cold season (1 November–31 March), it was −1.3 °C. The mean annual and cold season precipitation was 664 and 228 mm, respectively. The diversity of topography determines the differences in precipitation in the territory. The climate of Lithuania is formed by global teleconnection patterns and local climate factors that influence changes in the ice regime in rivers.
For ice phenological analysis, data from 24 water gauging stations (WGS) in 19 rivers of different sizes were selected (Table 1 and Figure 1). Daily ice cover data were extracted from the hydrological yearbooks of the Lithuanian Hydrometeorological Service (LHMS). Meteorological data (daily air temperature) were collected from 19 meteorological stations (MS) (Figure 1). Air temperature data for the period 1961–2020 were obtained from the LHMS database. The values of the five selected teleconnection patterns (TPs) were taken from the NOA database (https://www.cpc.ncep.noaa.gov/ (accessed on 30 November 2023)).

2.2. Methodology

When studying river ice phenology, three characteristics are described: freeze-up (date of ice cover appearance), break-up (date of ice cover disappearance), and ice cover duration. For the analysis, the duration of continuous ice cover was calculated, and freeze-up and break-up dates were determined for each year at all WGSs. Freeze-up and break-up dates were expressed as continuous days from 1 November. The freeze-up date is when a continuous ice cover is first detected at the water gauging point. The ice break-up date is the day when the continuous ice cover begins to break up, and the river does not freeze again. If a continuous ice cover has formed on the river more than once, that is, it froze and broke twice or more during the winter, the total duration of the ice cover was calculated by adding up the days of all periods with ice cover.
Long-term trends in air temperature and river ice characteristics through 1961–2020 were analyzed using the non-parametrical Mann–Kendall test, which is often applied to hydrometeorological data series [27]. The Mann–Kendall test evaluates whether the analyzed data series have an increasing or decreasing trend and determines their confidence level (p). If p < 0.05, the detected trend in the analyzed data series is statistically significant; otherwise (when p > 0.05), the trend is insignificant [28,29].
The Pettitt, SNHT, and Buishand tests are often used to check the homogeneity of analyzed hydrometeorological data series and to identify shift points [30]. The Pettitt test is a non-parametric method based on the Mann–Whitney and Wilcoxon tests to analyze the deviation of homogeneity in a time series [31]. In the SNHT, the values for a given year are compared with the average value of all data series [32]. The Buishand parametric test assumes the test values are independent and normally distributed [33]. The Buishand and Pettitt tests are more sensitive to shift points in the middle of the time series, while the SNHT is more sensitive to shift points near the beginning and end of the time series [34,35,36]. As with the Mann–Kendall test, the results are significant at p < 0.05.
When analyzing the change in hydrological parameters and to determine the underlying causes for their change, it is appropriate to compare the long-term series of these parameters with the TPs [37,38]. There are many TPs, so it makes sense to choose those that affect the climate of the Northern Hemisphere and especially the Baltic region. The North Atlantic Oscillation (NAO), Arctic Oscillation (AO), East Atlantic (EA), Scandinavia Pattern (SCAND), and Polar/Eurasia (POL–EUR) indices are commonly used to analyze European climate [39,40,41]. All these TPs are characterized by positive and negative phases followed by a certain atmospheric circulation. The NAO index is one of the most used in Europe. During the positive NAO phase, atmospheric pressure is higher than normal over the Azores and lower over Iceland. An increased atmospheric pressure gradient between Iceland and the Azores leads to stronger westerly transport in northern Europe, bringing warmer than normal weather. Meanwhile, during the negative phase of NAO, westerly transport weakens due to the reduced atmospheric pressure gradient, and then colder weather prevails more often in northern Europe [40]. The EA index is similar to the NAO and consists of north-south dipole anomaly centers spanning the North Atlantic from east to west. As in the case of the NAO, warmer-than-normal weather prevails in Europe during the positive EA phase and colder-than-normal during the negative phase. Very similar weather remains during the positive and negative AO index phases. In the case of the negative AO phase, a high-pressure area prevails in the polar region, and a low-pressure area prevails in the middle latitudes [42]. Therefore, during the negative AO phase, cold arctic air penetrates further into the southern latitudes, and therefore the air temperature is lower than usual [43]. During the positive phase of the AO, the high-pressure area in the mid-latitudes pushes the Arctic air further north and prevents it from spreading further, resulting in warmer-than-normal weather in Northern Europe [44]. The SCAND pattern is characterized by a high-pressure area over Scandinavia and a low-pressure area from western Europe to western Mongolia [45]. During the positive phase, the pressure over Scandinavia is higher; during the negative phase, it is lower than usual [46,47]. Because of this distribution of pressure fields, SCAND works in the opposite way to the previously discussed TPs. During its positive phase, winters are colder; during the negative phase, the winters are warmer than usual. Temperature anomalies in the eastern part of Asia and precipitation in north of Scandinavia are mostly associated with POL–EUR [48]. Therefore, it is likely that the signal of this index will be insignificant for the ice phenology of Lithuanian rivers.
Even though these TPs are widely used in the analysis of the climate of the Baltic region, it was initially necessary to find out whether there is a connection between them and the characteristics of ice cover of Lithuanian rivers. For this reason, in the first step, the Pearson correlations were estimated between (1) freeze-up dates and average values of TPs in Jul–Sep, Aug–Oct, Sep–Nov, Oct–Dec, Nov–Jan, Dec–Feb, Jan–Mar, and Feb–Apr months; (2) ice cover duration and average values of TPs in Oct–Jan, Nov–Feb, Dec–Mar, Jan–Apr, Feb–May, Nov–Mar, Nov–Apr, and Nov–May months; and (3) break-up dates and average values of TPs in Oct–Dec, Nov–Jan, Dec–Feb, Jan–Mar, Feb–Apr, and Mar–May months. In the second step, the three strongest correlations between ice cover characteristics and TPs were selected, and the Wilcoxon rank sum test was used to determine whether there was a signal from TP to ice characteristics. The non-parametric Wilcoxon rank sum test is based on comparing two independent data samples. It determines whether the null hypothesis is accepted (two independent samples are statistically similar) or rejected (two independent samples are statistically different) [49]. Test results are given with their confidence levels (p 0.1, 0.05, 0.01).

3. Results

3.1. River Ice Phenology Changes in Lithuania during Two Climate Normal Periods

Based on the data of daily observations, the average dates of river freeze-up and ice break-up in the selected first and second climate normal periods were calculated and compared (Figure 2). This analysis revealed that the freezing dates changed the least: in the first 30-year period, it was at the beginning of January, while in the second period, it was observed in the middle of January. Despite this, the freezing dates in individual rivers differed significantly. For example, in the first period, the earliest ice formed in early December in the river Šuš-Šia and the latest in early February in Ner-Vil. In the second period, the earliest ice formed in the middle of December in Dub-Lyd and the latest in the second decade of February in Ner-Vil. As we can see from Figure 2, in the first period, in most of the rivers, the ice broke up on average in early March. However, in the later period, this event occurred a month earlier and started at the beginning of February. A comparison of these ice regime phases in different periods of climate normal indicated that river freezing started on average one week later in the second period (1991–2020), and ice broke up three weeks earlier than in the first period (1961–1990).
Another important characteristic of ice is the duration of its cover. As shown in Figure 3, ice phenomena began to decrease remarkably already in the first period (1961–1990) of the climate normal. In 1961–1970, with few exceptions, ice still covered all studied rivers for an average of almost 74 days. However, between 1971 and 1980, it gradually began to diminish and was nearly halved to 40 days. In the last decade of the first 30-year period, ice cover was observed only for an average of 14 days. It has almost disappeared in such relatively large rivers as the Nemunas, Šventoji, and Žeimena (Figure 3). In the second period (1991–2020) of the climate normal, the loss of river ice cover accelerated even more. In the first decade, it was present for an average of 19 days; in the second, it averaged 12 days; and in the last, it lasted for only six days. An extremely cold winter of 1995–1996 stood out, during which the sum of negative air temperatures exceeded 800 °C, and the average duration of the ice cover was 71 days. Such cold winters were typical only at the beginning of the first climate normal, when below-zero temperatures persisted for a long time, and ice covered the rivers for 70 days or longer.

3.2. Air Temperature Changes during Two Climate Normal Periods

Since river ice phenology is closely related to air temperature, the analysis of this local climate factor was initially conducted in two climate normal periods. Data from 19 meteorological stations (Biržai MS, Dotnuva MS, Dūkštas MS, Kaunas MS, Kybartai MS, Klaipėda MS, Laukuva MS, Lazdijai MS, Nida MS, Palanga MS, Panevėžys MS, Raseiniai MS, Šiauliai MS, Šilutė MS, Telšiai MS, Ukmergė MS, Utena MS, Varėna MS, and Vilnius MS) from the entire territory of Lithuania were used. Average annual and sum of negative air temperatures in November–March are presented in Figure 4. As the graph indicates, the average annual air temperature tended to increase, while the sum of negative air temperatures, on the contrary, tended to decrease throughout the entire studied period. The average annual air temperature during the first period of climate normal (1961–1990) was 6.2 °C and rose by 0.03 °C per year. In the second period (1991–2020), it was 1.2 °C higher and increased by 0.05 °C annually. After summarizing the data for both thirty-year periods, it was found that the average annual air temperature got higher by 0.04 °C per year. The Mann–Kendall test revealed a statistically insignificant (p = 0.37) increase in the average annual air temperature in the first period of the climate normal. However, in the second period and throughout 1960–2020, the positive trends were statistically significant (p 0.002 and <0.0001, respectively). Similar trends were identified in the case of the sum of negative air temperatures. It decreased by an average of 5.5 °C per year throughout the study period, which was statistically significant (p = 0.001). In the first period, its average value was 493.6 °C, decreasing by 10.5 °C per year, while in the second period, it reached only 332.3 °C on average and decreased by 4.6 °C per year. Although the sum of below-zero temperatures in both analysed periods tended to decline, a statistically significant negative trend was found only in the first period (p = 0.04).
To find out whether this division of the study period (1960–2020) into two equal subperiods for the present analysis was correct, homogeneity tests were applied. The Pettitt, SNHT, and Buishand tests were chosen. Based on them, it was established that the breaking point in the series of average annual air temperatures was in 1988, and in the series of the sum of negative air temperatures in 1986 (Table 2). These two dates are very close to the end of the first climate normal, so this division not only helped to estimate the differences between the studied 30-year periods but also almost coincided with the boundaries of the climate normal periods.

3.3. Relationship between River Ice Phenology and Air Temperature Regime in Lithuania

After analyzing the parameters of ice phenology, their dependence on air temperature was studied. The sum of negative air temperatures from November 1 to the freezing date was used to analyze the dates of ice formation. Meanwhile, the sum of negative air temperatures from November to March was used to analyze ice cover duration and ice break-up dates. The average air temperature was calculated using the data from meteorological stations located in the studied river catchments. The closest (negative) relationship was found between the sum of negative air temperatures before freezing and the date of freezing (Appendix A). Based on data from 24 WGSs, where observations of ice phenomena were carried out, the average correlation was found to be 0.71 and ranged from 0.40 (Ner-Vil) to 0.88 (Nem-Nem) in individual cases. A slightly weaker relationship was established between the sum of negative air temperatures and ice cover duration. This ratio reached an average of 0.62, but in individual cases varied from 0.45 (Jūr-Tau) to 0.79 (Svy-Gun and Lėv-Ber). Although the linear relationship between these two parameters was not very strong, based on the graphs in Figure 5, it was reasonably reliable. In the second period of the climate normal, the sum of negative air temperatures was noticeably lower than in the first. Similar trends were also observed when analyzing the duration of ice cover; i.e., in the second period, it was significantly shorter than in the first. Summing up the results of all 24 studied rivers, it was estimated that in one river, ice cover duration decreased by up to 25%, in eight rivers by 25 to 50%, in thirteen rivers by 50 to 75%, and in two rivers by more than 75%. Therefore, in the majority of rivers, the length of the ice season has shrunk by 25 to 75%. The duration of the ice cover has shortened the least in Ven-Lec and the most in Nev-Pan and Jūr-Tau.
The weakest relationship was identified between the sum of negative air temperatures and ice break-up dates (Appendix A). Even though the correlation reached 0.77–0.83 in individual rivers, on average it was only 0.58.

3.4. Relationship between River Ice Phenology and Northern Hemisphere Teleconnection Patterns (TPs)

One of the tasks of this study was to determine the relationship between the ice regime of Lithuanian rivers (dates of freezing and ice break-up and duration of ice cover) and Northern Hemisphere TPs such as the North Atlantic Oscillation (NAO), Arctic Oscillation (AO), East Atlantic (EA), Polar/Eurasia (POL), and Scandinavia (SCAND).
The weakest relationships were found between freezing date and average TP value in selected months (Table 3). Despite the higher correlation (up to 0.70) between individual rivers and TPs, the average coefficient for all rivers was very low. The strongest relationship was determined between the freezing date and the EA average in August–October (0.26) and the NAO in November–January (0.24), while with other indices the coefficient did not reach 0.20 (Table 3).
A considerably stronger relationship was detected between ice cover duration, break-up dates, and the TPs. To analyze the long-term changes in the duration of the ice cover, the relationship between this parameter and the average value of TP in November–February, November–March, and November–April was investigated (Table 4). The length of ice season was visibly linked to the NAO and AO indices in November–March. Even though the average correlation of the selected rivers was relatively low (for NAO −0.59, for AO −0.48), it was up to −0.83 in individual cases. A weaker relationship was found between ice cover duration and the EA, SCAND, and POL–EUR (−0.39, 0.31, and 0.31, respectively).
Analysis of the dependence of ice break-up dates on TPs (Table 5) revealed similar correlation coefficients as in the case of ice cover duration. As indicated in Table 5, the ice break-up dates were best correlated with the average values of NAO and AO in January–March (up to −0.54 and −0.44, respectively), while the relation with other TPs was significantly weaker.
Next, a non-parametric Wilcoxon rank sum test was applied to determine whether the TP signal detected for ice phenological parameters was significant. Because this test requires a long data series, only those rivers with the longest ice cover persistence were selected. As in the case of linear correlation, the Wilcoxon rank sum test revealed that the signal between the average TP values in individual months and freeze-up dates was significant only in some rivers. The EA index stood out the most; its signal was determined on freeze-up dates in all studied rivers, with α ranging from 0.1 to 0.01. The significance of this signal was highest when applying the EA average for July–September, lower for September–November, and the lowest for August–October (Table 6). The analysis of individual rivers showed that the weakest signal was on Ver-Ver and Svy-Gun, and the strongest on Šus-Šia, Šuš-Jos, and Mūš-Ust. Less significant signals were recorded between the AO and NAO and freeze-up date. Significant signals were found for the AO in five out of nine cases and for the NAO in four out of nine cases, with α ranging from 0.1 to 0.01. Meanwhile, only isolated and not always high-significance signals were detected between the POL–EUR and SCAND and freeze-up dates, suggesting that they were relatively random.
Table 7 shows that the strongest signals were between all fitted TPs and the duration of the ice cover. The signal of the average NAO value in November–March for the length of ice season was particularly strong, with a significance level of 0.01 in eight out of nine cases. The second strongest signal was of the EA in November–April, and the third was the AO in November–February, at the 0.01 significance level in six out of nine cases and four out of nine, respectively. Weak signals were recorded between the SCAND and POL–EUR and ice cover duration. The SCAND signal was most significant when evaluating its value in November–March and the POL–EUR signal was the most significant in terms of its value in November–April.
Analysis of the signals between TPs and ice break-up dates revealed that their nature was similar to that of ice cover duration (Table 8). The average value of NAO in January–March had the strongest influence on the break-up date (at the significance level of 0.01). According to the signals’ significance level, the EA and AO were slightly behind. For the AE, the strongest signal was obtained between its average value for December–February (the confidence level was 0.01 in six out of nine cases, and 0.05 in the remaining three out of nine) and the break-up date. Meanwhile, the AO signal for break-up dates was strongest when using the values of this index for January–March (the confidence level was 0.01 in four cases out of nine, and 0.05 in the rest). According to the significance of the signals, the SCAND was in the penultimate place, and the POL–EUR was in last place when estimating their average values for February–April.
Due to TPs regularities, in the case of negative values of NAO, EA, and AO indices, rivers should freeze earlier, and their ice break up later than during positive phases. Meanwhile, in the case of SCAND, on the contrary, the ice cover should form earlier and disappear later when the values of this index become positive. Similar regularities were also determined when analyzing the ice phenology regime of Lithuanian rivers. However, these regularities were confirmed only in the cases of NAO, EA, and AO circulations, and only when the confidence level of these TPs signals to the dates of freeze-up and break-up as well as the duration of river ice cover was 0.05 and higher. The patterns of the SCAND type in Lithuanian rivers were determined only by analyzing the duration of ice cover and break-up date, and no reliable signals were found in the case of freezing date. Meanwhile, the POL–EUR connection with the ice phenology of Lithuanian rivers was the weakest and determined only episodically.

4. Discussion

This study aimed to determine changes in the ice phenology of Lithuanian rivers. The results revealed that trends in river ice freeze-up and break-up dates and ice cover duration are consistent with those in other Northern Hemisphere rivers and are a direct consequence of ongoing climate change.
The performed analysis was distinguished by the fact that it examined changes over two periods of climate normal and included the most recent 30-year period, recommended as a standard reference for long-term climate change assessments [25]. Comparing the periods of 1961–1990 and 1991–2020, we unveiled considerable phenological changes in the ice regime of Lithuanian rivers. It was found that, recently, the average freeze-up date was delayed by eight days, the ice-breaking date occurred 24 days earlier, and the duration of the ice cover was shortened by half (from 67 to 34 days). Unfortunately, no similar studies could be found that assessed ice processes over the same periods. Therefore, the authors of this study suggest and encourage the application of the used methodology and the chosen periods to other climate-related studies.
Although the trends in ice phenology parameters are consistent, loss rates vary across countries. A similar study in neighbouring Latvia, based on ice data from 1945 to 2012, demonstrated that the duration of ice cover on rivers decreased by 6–15 days on average and concluded that ice-free winters have been becoming more frequent since the 1970s [17]. Since the second half of the 20th century, ice cover has also rarely been observed in the lower part of the Danube River [9]. In the last 60 years, there have been only ten winters when this second-largest European river has been covered with ice. In Sweden, in 1985–2014, compared to the previous three decades, the average duration of ice cover decreased by 11–28 days, with the freeze-up occurring on average about 10 days later and the break-up approximately 17 days earlier in southern Sweden [16]. In addition, in this Scandinavian country, there was an increase in cases of very short ice cover duration (i.e., less than 50 days). In Mongolia, river freeze-up and break-up dates have shifted by 3–30 days over the last 60 years [14].
Many studies have highlighted the sensitivity of river ice to climate change, especially to rising air temperatures. Indeed, milder winter conditions have a decisive influence on the presence of river ice [10,14,17,50]. Our study found that the sum of the negative air temperatures of the cold season (from 1 November to 31 March) is a good predictor of the duration of river ice cover, and ice break-up and freeze-up dates (in this case, the air temperatures were summed up to this date). These findings align with others [17,51,52]. Although our study was limited to air temperature data, there is evidence of other climate factors that may influence the investigated ice variables. Different forms of precipitation: snow or rain (an influence of air temperature, albeit indirect) can affect river ice on/off processes in various phases of ice sheet formation. In spring, ice break-up can be delayed due to snowfall and occur earlier due to rainfall [50,53].
Like other climate variables, air temperature varies not only in time but also in space. This is why many studies point to different responses of the ice regime to climate change in different regions. Some studies show that the rate of phenological changes in river ice intensifies southward; e.g., the authors of [16] found that the rate of change in ice freeze-up and break-up dates was roughly twice as high in southern Sweden as in the northern part. Other studies [17,54] suggest that these changes are more pronounced in rivers closer to the sea. As climate changes, the geographical ice margin is expected to move northward [11,55]. Nonetheless, the present study did not demonstrate regional patterns in river ice phenology. A possible explanation could be the great diversity of the studied rivers. Individual river features can also significantly affect the river ice formation process [10,56]. In addition to those mentioned dominant climatic factors, other drivers such as river hydrogeomorphology (water level, discharge, size, sinuosity, elevation, etc.) are often discussed [7,56,57]. Due to these characteristics, small rivers are distinguished as more sensitive (in terms of ice regime) to the variability of weather events in the cold period as their water loses or gains its heat content faster [50,58,59]. Therefore, further studies that consider not only climate indicators are needed to understand and predict changes in complex river ice phenomena.
The study revealed that trends in ice phenology matched not only air temperature trends but also teleconnection patterns. Of the five large-scale drivers studied, the North Atlantic Oscillation (NAO) had the greatest influence on ice cover duration and ice break-up dates, and the river freeze-up process was best correlated with the East Atlantic (EA) climate pattern. The Polar/Eurasia (POL–EUR) pattern was found to be the least related to the ice phenology of Lithuanian rivers. The relation between the NAO index and ice break-up dates in the lower reaches of the Nemunas River was determined [20]. The effect of this index on the variability of the ice regime and winter conditions in general in Latvia and the whole Baltic region was documented [54,60]. However, a significant relationship between the NAO index and the ice regime cannot always be detected [9,61]. The negative phase of the EA is also widely known to be associated with low winter temperature anomalies in Europe [62].
Many of the studies cited in this paper confirm dramatic changes in river ice regime over the past 30 years (1990–2020). Findings suggest that climate change is making this complex phenomenon increasingly rare, and there is evidence that it may disappear permanently in some areas [9]. Meanwhile, even thinner ice cover (due to a warming climate) can reduce the threat of ice jamming [55]. Therefore, there is a probability that in ice-affected rivers, the interaction of river ice with infrastructures such as riverside buildings, bridges, culverts, dykes, etc., will occur less frequently. Thus, adding a climate-change component in designing new structures or revisiting the existing ones to ascertain their safety should be considered [5].
If the established river ice trends do not change, similar research will not be possible in 20–30 years. The uniqueness of this study lies not only in the fact that it examines the significantly shrinking river ice cover; unfortunately, the availability of river ice data is decreasing not only due to natural reasons but also because there are fewer and fewer traditional gauging stations remaining. River ice cover is changing rapidly, and this will be accompanied by a series of changes in riverine ecosystems that will affect essential ecosystem services and, more broadly, the sustainability of crucial resources, livelihoods, and traditional practices [55]. The knowledge gained implies that such research (the current study) can help us understand, adapt to, and prepare for anticipated changes. Unfortunately, this unique phase of the hydrological regime of rivers remains underestimated despite its negative and positive effects on the environment, society, and the economy, despite objective and less objective reasons [8,56,57]. Scientists are concerned that significantly reduced ice monitoring systems, uncertainties, and gaps in knowledge about river ice phenomena make this research more difficult [7,61,63,64].

5. Conclusions

  • The study has found considerable changes in the characteristics of river ice cover during two periods of the climate normal (1961–1990 and 1991–2020). In the second period, the rivers froze on average nine days later, and their ice broke up 24 days earlier than in the first period. As a result, the duration of the ice cover has decreased by 33 days in the recent period.
  • Significant trends in air temperature increase have also been determined during the 60-year period. This has been particularly evident in the last 30-year period, when temperatures have increased by 0.5 °C per decade, compared to just 0.3 °C in the first 30 years. A negative linear correlation was obtained between air temperature and river ice parameters (from −0.34 to −0.88); therefore, it can be concluded that the loss of ice cover on rivers was influenced by the increased air temperature.
  • The analysis of freezing dates showed that they were significantly influenced by the EA index. Meanwhile, the ice cover duration and the break-up dates were affected by the NAO, whose signal was significant in all the studied cases. The POL–EUR and SCAND indices also influenced river ice characteristics, but their significant signals were determined only for individual rivers.
  • Since the phenomenon of river ice is of great importance for nature and humans (due to ecosystem services), its changes will affect many aspects of ecosystem functioning and human well-being. The insights gained from this research may be of assistance to policymakers and general readers who may still have doubts about the phenomenon of climate change.

Author Contributions

Conceptualization, D.Š., D.J., A.J. and J.K.; methodology, D.Š., D.J., A.J. and J.K.; data collection, A.J.; formal analysis, D.Š., D.J., A.J. and J.K.; writing—original draft preparation, D.Š., D.J., A.J. and J.K.; writing—review and editing, D.Š., D.J., A.J. and J.K.; visualization D.J. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Acknowledgments

The authors wish to thank the Lithuanian Hydrometeorological Service for providing the daily hydrometeorological data.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Correlation between ice parameters and air temperature in the selected rivers.
Table A1. Correlation between ice parameters and air temperature in the selected rivers.
Correlation betweenNem-DruNem-NemNem-SmaVer-VerNer-VilNe-JonŽei-PabŠve-UkmNev-PanŠuš-ŠiaŠuš-JosDub-LydJūr-TauAkm-PaaŠeš-SkiMin-KarSvy-GunNem-TabMūš-UstLėv-BerTat-TrečVen-PapVen-LecBar-SkuAverage
Freeze-up date and TSumNeg 1−0.73−0.88−0.78−0.61−0.40−0.87−0.68−0.79−0.67−0.79−0.76−0.75−0.81−0.74−0.85−0.76−0.61−0.81−0.65−0.67−0.77−0.72−0.80−0.80−0.74
Duration and TSumNeg 2−0.56−0.57−0.62−0.71−0.53−0.48−0.57−0.57−0.51−0.68−0.66−0.60−0.45−0.65−0.58−0.57−0.79−0,62−0.73−0.79−0.73−0.77−0.64−0.54−0.62
Break-up date and TSumNeg 3−0.45−0.70−0.63−0.49−0.53−0.34−0.83−0.56−0.42−0.59−0.53−0.45−0.77−0.43−0.72−0.74−0.64−0.59−0.63−0.62−0.67−0.66−0.36−0.71−0.59
1 Correlation between the date of freezing and the sum of negative air temperature before the freezing. 2 Correlation between the ice cover duration and the sum of negative air temperatures in November–March. 3 Correlation between the break-up date and the sum of negative air temperatures in November–March.

References

  1. Tuckett, R. Greenhouse Gases and the Emerging Climate Emergency. In Climate Change; Elsevier: Amsterdam, The Netherlands, 2021; pp. 19–45. [Google Scholar] [CrossRef]
  2. Allen, M.R.; Dube, O.P.; Solecki, W.; Aragón-Durand, F.; Cramer, W.; Humphreys, S.; Kainuma, M.; Kala, J.; Mahowald, N.; Mulugetta, Y.; et al. Framing and Context. In Global Warming of 1.5 °C. An IPCC Special Report on the Impacts of Global Warming of 1.5 °C above Pre-Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change, Sustainable Development, and Efforts to Eradicate Poverty; Masson-Delmotte, V., Zhai, P., Pörtner, H.-O., Roberts, D., Skea, J., Shukla, P.R., Pirani, A., Moufouma-Okia, W., Péan, C., Pidcock, R., et al., Eds.; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2018; pp. 49–92. [Google Scholar] [CrossRef]
  3. WMO. WMO Global Annual to Decadal Climate Update; World Meteorological Organization: Geneva, Switzerland, 2023; p. 24. Available online: https://hadleyserver.metoffice.gov.uk/wmolc/WMO_GADCU_2023-2027.pdf (accessed on 13 November 2023).
  4. Bennett, K.E.; Prowse, T.D. Northern Hemisphere Geography of Ice-covered Rivers. Hydrol. Process. 2009, 24, 235–240. [Google Scholar] [CrossRef]
  5. Beltaos, S.; Burrell, B.C. Climatic Change and River Ice Breakup. Can. J. Civ. Eng. 2003, 30, 145–155. [Google Scholar] [CrossRef]
  6. Yang, X.; Pavelsky, T.M.; Allen, G.H. The Past and Future of Global River Ice. Nature 2020, 577, 69–73. [Google Scholar] [CrossRef] [PubMed]
  7. Prowse, T.D.; Bonsal, B.R.; Duguay, C.R.; Lacroix, M.P. River-Ice Break-up/Freeze-up: A Review of Climatic Drivers, Historical Trends and Future Predictions. Ann. Glaciol. 2007, 46, 443–451. [Google Scholar] [CrossRef]
  8. Beltaos, S.; Prowse, T. River-ice Hydrology in a Shrinking Cryosphere. Hydrol. Process. 2008, 23, 122–144. [Google Scholar] [CrossRef]
  9. Ionita, M.; Badaluta, C.-A.; Scholz, P.; Chelcea, S. Vanishing River Ice Cover in the Lower Part of the Danube Basin—Signs of a Changing Climate. Sci. Rep. 2018, 8, 7948. [Google Scholar] [CrossRef] [PubMed]
  10. Graf, R. Estimation of the Dependence of Ice Phenomena Trends on Air and Water Temperature in River. Water 2020, 12, 3494. [Google Scholar] [CrossRef]
  11. Alfredsen, K.; Bridges, R.; Hendrikse, H.; Høyland, K.V.; Kolerski, T.; Leppäranta, M.; Peng, L.; Guo, X. How does climate change affect ice formation and presence in rivers, lakes and oceans, as well as its impact on infrastructure. In Hydrolink 2022/3; International Association for Hydro-Environment Engineering and Research (IAHR): Madrid, Spain, 2022; pp. 77–79. Available online: https://www.iahr.org/library/infor?pid=21855 (accessed on 14 November 2023).
  12. Magnuson, J.J.; Robertson, D.M.; Benson, B.J.; Wynne, R.H.; Livingstone, D.M.; Arai, T.; Assel, R.A.; Barry, R.G.; Card, V.; Kuusisto, E.; et al. Historical Trends in Lake and River Ice Cover in the Northern Hemisphere. Science 2000, 289, 1743–1746. [Google Scholar] [CrossRef]
  13. Newton, A.M.W.; Mullan, D.J. Climate Change and Northern Hemisphere Lake and River Ice Phenology from 1931–2005. Cryosphere 2021, 15, 2211–2234. [Google Scholar] [CrossRef]
  14. Punsalmaa, B.; Nyamsuren, B.; Buyndalai, B. Trends in River and Lake Ice in Mongolia. Assessments of Impacts and Adaptations to Climate Change (AIACC); Working Paper 4. 2004. Available online: http://www.start.org/Projects/AIACC_Project/working_papers/Working%20Papers/AIACC_WP_No004.pdf (accessed on 30 November 2023).
  15. Vuglinsky, V.; Valatin, D. Changes in Ice Cover Duration and Maximum Ice Thickness for Rivers and Lakes in the Asian Part of Russia. Nat. Resour. 2018, 9, 73–87. [Google Scholar] [CrossRef]
  16. Hallerbäck, S.; Huning, L.S.; Love, C.; Persson, M.; Stensen, K.; Gustafsson, D.; AghaKouchak, A. Climate Warming Shortens Ice Durations and Alters Freeze and Break-up Patterns in Swedish Water Bodies. Cryosphere 2022, 16, 2493–2503. [Google Scholar] [CrossRef]
  17. Latkovska, I.; Apsīte, E.; Elferts, D. Long-Term Changes of the Ice Regime of Rivers in Latvia. Hydrol. Res. 2016, 47, 782–798. [Google Scholar] [CrossRef]
  18. Majewski, W. Ice Phenomena on the Lower Vistula. Geophysica 2011, 47, 57–67. [Google Scholar]
  19. Sharma, S.; Magnuson, J.J.; Batt, R.D.; Winslow, L.A.; Korhonen, J.; Aono, Y. Direct Observations of Ice Seasonality Reveal Changes in Climate over the Past 320–570 Years. Sci. Rep. 2016, 6, 25061. [Google Scholar] [CrossRef] [PubMed]
  20. Stonevicius, E.; Stankunavicius, G.; Kilkus, K. Ice Regime Dynamics in the Nemunas River, Lithuania. Clim. Res. 2008, 36, 17–28. [Google Scholar] [CrossRef]
  21. IPCC. IPCC Special Report on the Ocean and Cryosphere in a Changing Climate; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2019; p. 755. [Google Scholar]
  22. von Storch, H.; Omstedt, A.; Pawlak, J.; Reckermann, M. Introduction and Summary. In Second Assessment of Climate Change for the Baltic Sea Basin; The BACC II Author Team, Ed.; Springer: Cham, Switzerland, 2015; pp. 1–22. [Google Scholar] [CrossRef]
  23. Klavins, M.; Briede, A.; Rodinov, V.; Frisk, T. Ice Regime of Rivers in Latvia in Relation to Climatic Variability. Int. Ver. Theor. Angew. Limnol. Verhandlungen 2006, 29, 1825–1828. [Google Scholar] [CrossRef]
  24. Šarauskienė, D.; Jurgelėnaitė, A. Impact of Climate Change on River Ice Phenology in Lithuania. Environ. Res. Eng. Manag. 2008, 46, 13–22. [Google Scholar]
  25. WMO. WMO Guidelines on the Calculation of Climate Normals; World Meteorological Organization: Geneva, Switzerland, 2017; p. 29. [Google Scholar]
  26. Lithuanian Hydrometeorological Service under the Ministry of Environment. Climate Atlas of Lithuania; Petro Ofsetas: Vilnius, Lithuania, 2013; p. 175. [Google Scholar]
  27. Hu, Z.; Liu, S.; Zhong, G.; Lin, H.; Zhou, Z. Modified Mann-Kendall Trend Test for Hydrological Time Series under the Scaling Hypothesis and Its Application. Hydrolog. Sci. J. 2020, 65, 2419–2438. [Google Scholar] [CrossRef]
  28. Mann, H.B. Nonparametric Tests Against Trend. Econometrica 1945, 13, 163–171. [Google Scholar] [CrossRef]
  29. Kendall, M.G. Rank Correlation Methods; Oxford University Press: Oxford, UK, 1975. [Google Scholar]
  30. Yildirim, G.; Rahman, A. Homogeneity and Trend Analysis of Rainfall and Droughts over Southeast Australia. Ant. Hazards 2022, 112, 1657–1683. [Google Scholar] [CrossRef]
  31. Pettitt, A.N. A Non-Parametric Approach to the Change-Point Problem. Appl. Stat. 1979, 28, 126–135. [Google Scholar] [CrossRef]
  32. Alexandersson, H. A Homogeneity Test Applied to Precipitation Data. J. Climatol. 1986, 6, 661–675. [Google Scholar] [CrossRef]
  33. Buishand, T.A. Some Methods for Testing the Homogeneity of Rainfall Records. J. Hydrol. 1982, 58, 11–27. [Google Scholar] [CrossRef]
  34. Hawkins, D.M. Testing a Sequence of Observations for a Shift in Location. J. Am. Stat. Assoc. 1977, 72, 180–186. [Google Scholar] [CrossRef]
  35. Wijngaard, J.B.; Klein Tank, A.M.G.; Können, G.P. Homogeneity of 20th Century European Daily Temperature and Precipitation Series. Int. J. Climatol. 2003, 23, 679–692. [Google Scholar] [CrossRef]
  36. Costa, A.C.; Soares, A. Homogenization of Climate Data: Review and New Perspectives Using Geostatistics. Math. Geosci. 2009, 41, 291–305. [Google Scholar] [CrossRef]
  37. Stephenson, D.B.; Wanner, H.; Brönnimann, S.; Luterbacher, J. The History of Scientific Research on the North Atlantic Oscillation. Geophys. Monogr. Ser. 2003, 134, 37–50. [Google Scholar] [CrossRef]
  38. Rousi, E.; Rust, H.W.; Ulbrich, U.; Anagnostopoulou, C. Implications of Winter NAO Flavors on Present and Future European Climate. Climate 2020, 8, 13. [Google Scholar] [CrossRef]
  39. Plewa, K.; Perz, A.; Wrzesiński, D. Links between Teleconnection Patterns and Water Level Regime of Selected Polish Lakes. Water 2019, 11, 1330. [Google Scholar] [CrossRef]
  40. Jiménez-Guerrero, P.; Ratola, N. Influence of the North Atlantic Oscillation on the Atmospheric Levels of Benzo[a]Pyrene over Europe. Clim. Dyn. 2021, 57, 1173–1186. [Google Scholar] [CrossRef]
  41. Meier, H.E.M.; Kniebusch, M.; Dieterich, C.; Gröger, M.; Zorita, E.; Elmgren, R.; Myrberg, K.; Ahola, M.P.; Bartosova, A.; Bonsdorff, E.; et al. Climate Change in the Baltic Sea Region: A Summary. Earth Syst. Dynam. 2022, 13, 457–593. [Google Scholar] [CrossRef]
  42. Gečaitė, I.; Rimkus, E. Snow Cover Regime in Lithuania. Geography 2010, 46, 17–24, (In Lithuanian with English summary). [Google Scholar]
  43. Gecaite, I.; Pogoreltsev, A.; Ugryumov, A. Arctic Oscillation Impact on Thermal Regime of the Baltic Region Eastern Part. Solnechno-Zemn. Fiz. 2016, 2, 64–70. [Google Scholar] [CrossRef] [PubMed]
  44. Thompson, D.W.J.; Wallace, J.M. The Arctic Oscillation signature in wintertime geopotential height and temperature fields. Geophys. Res. Lett. 1998, 25, 1297–1300. [Google Scholar] [CrossRef]
  45. Barnston, A.G.; Livezey, R.E. Classification, Seasonality and Persistence of Low-Frequency Atmospheric Circulation Patterns. Mon. Wea. Rev. 1987, 115, 1083–1126. [Google Scholar] [CrossRef]
  46. Liu, Y.; Wang, L.; Zhou, W.; Chen, W. Three Eurasian Teleconnection Patterns: Spatial Structures, Temporal Variability, and Associated Winter Climate Anomalies. Clim. Dyn. 2014, 42, 2817–2839. [Google Scholar] [CrossRef]
  47. Nojarov, P. The Increase in September Precipitation in the Mediterranean Region as a Result of Changes in Atmospheric Circulation. Meteorol. Atmos. Phys. 2016, 129, 145–156. [Google Scholar] [CrossRef]
  48. Gao, N.; Bueh, C.; Xie, Z.; Gong, Y. A Novel Identification of the Polar/Eurasia Pattern and Its Weather Impact in May. J. Meteorol. Res. 2019, 33, 810–825. [Google Scholar] [CrossRef]
  49. Fay, M.P.; Proschan, M.A. Wilcoxon-Mann-Whitney or t-Test? On Assumptions for Hypothesis Tests and Multiple Interpretations of Decision Rules. Statist. Surv. 2010, 4, 1–39. [Google Scholar] [CrossRef]
  50. Chen, Y.; She, Y. Long-Term Variations of River Ice Breakup Timing across Canada and Its Response to Climate Change. Cold Reg. Sci. Technol. 2020, 176, 103091. [Google Scholar] [CrossRef]
  51. Pawłowski, B. Long-term Variability in the Course of Ice Phenomena on the Vistula River in Toruń. Bull. Geogr. Phys. Geogr. Ser. 2009, 1, 91–102. [Google Scholar] [CrossRef]
  52. Agafonova, S.A.; Frolova, N.A.; Surkova, G.V.; Koltermann, K.P. Modern Characteristics of the Ice Regime of Russian Arctic Rivers and their Possible Changes in the 21st Centure. Geogr. Environ. Sustain. 2017, 10, 4–15. [Google Scholar] [CrossRef]
  53. Lesack, L.F.W.; Marsh, P.; Hicks, F.E.; Forbes, D.L. Timing, Duration, and Magnitude of Peak Annual Water-Levels during Ice Breakup in the Mackenzie Delta and the Role of River Discharge. Water Resour. Res. 2013, 49, 8234–8249. [Google Scholar] [CrossRef]
  54. Klavins, M.; Briede, A.; Rodinov, V. Long Term Changes in Ice and Discharge Regime of Rivers in the Baltic Region in Relation to Climatic Variability. Clim. Chang. 2009, 95, 485–498. [Google Scholar] [CrossRef]
  55. Burrell, B.C.; Beltaos, S.; Turcotte, B. Effects of Climate Change on River-Ice Processes and Ice Jams. Int. J. River Basin Manag. 2022, 21, 421–441. [Google Scholar] [CrossRef]
  56. Chu, T.; Das, A.; Lindenschmidt, K.-E. Monitoring the Variation in Ice-Cover Characteristics of the Slave River, Canada Using RADARSAT-2 Data-A Case Study. Remote Sens. 2015, 7, 13664–13691. [Google Scholar] [CrossRef]
  57. Thellman, A.; Jankowski, K.J.; Hayden, B.; Yang, X.; Dolan, W.; Smits, A.P.; O’Sullivan, A.M. The Ecology of River Ice. J. Geophys. Res. Biogeosci. 2021, 126, e2021JG006275. [Google Scholar] [CrossRef]
  58. Buffin-Bélanger, T.; Bergeron, N.; Dubé, J. Ice Formation in Small Rivers. In River Ice Formation. Committee on River Ice Processes and the Environment; CGU-HS: Edmonton, AB, Canada, 2013; pp. 385–409. [Google Scholar]
  59. Lind, L.; Alfredsen, K.; Kuglerová, L.; Nilsson, C. Hydrological and Thermal Controls of Ice Formation in 25 Boreal Stream Reaches. J. Hydrol. 2016, 540, 797–811. [Google Scholar] [CrossRef]
  60. Klavins, M.; Avotniece, Z.; Rodinovs, V. Dynamics and Impacting Factors of Ice Regimes in Latvia Inland and Coastal Waters. Proc. Latv. Acad. Sci. Sect. B Nat. Exact Appl. Sci. 2016, 70, 400–408. [Google Scholar] [CrossRef]
  61. Gebre, S.B.; Alfredsen, K.T. Investigation of river ice regimes in some Norwegian water courses. In Proceedings of the 16th Workshop on the Hydraulics of Ice Covered Rivers, Winnipeg, MB, Canada, 18–22 September 2011; CGU HS Committee on River Ice Processes and the Environment: Winnipeg, MB, Canada, 2011; pp. 1–20. [Google Scholar]
  62. Mikhailova, N.V.; Yurovsky, A.V. The East Atlantic Oscillation: Mechanism and Impact on the European Climate in Winter. Phys. Oceanogr. 2016, 4, 25–33. [Google Scholar] [CrossRef]
  63. Lindenschmidt, K.-E.; Baulch, H.; Cavaliere, E. River and Lake Ice Processes—Impacts of Freshwater Ice on Aquatic Ecosystems in a Changing Globe. Water 2018, 10, 1586. [Google Scholar] [CrossRef]
  64. Derksen, C.; Burgess, D.; Duguay, C.; Howell, S.; Mudryk, L.; Smith, S.; Thackeray, C.; Kirchmeier-Young, M. Changes in snow, ice, and permafrost across Canada. In Canada’s Changing Climate Report; Bush, E., Lemmen, D.S., Eds.; Government of Canada: Ottawa, ON, Canada, 2019; pp. 194–260. [Google Scholar]
Sustainability 16 00725 g001
Figure 1. Study area and measurement stations.
Figure 1. Study area and measurement stations.
Sustainability 16 00725 g001
Sustainability 16 00725 g002
Figure 2. The average freeze-up and break-up dates of Lithuanian rivers in the different periods (abbreviations of river names are provided in Table 1).
Figure 2. The average freeze-up and break-up dates of Lithuanian rivers in the different periods (abbreviations of river names are provided in Table 1).
Sustainability 16 00725 g002
Sustainability 16 00725 g003
Figure 3. The duration of ice cover of Lithuanian rivers in 1961–2020.
Figure 3. The duration of ice cover of Lithuanian rivers in 1961–2020.
Sustainability 16 00725 g003
Sustainability 16 00725 g004
Figure 4. Average annual (a) and sum of negative air temperatures (b) in November–March according to the data of 19 meteorological stations in Lithuania.
Figure 4. Average annual (a) and sum of negative air temperatures (b) in November–March according to the data of 19 meteorological stations in Lithuania.
Sustainability 16 00725 g004
Sustainability 16 00725 g005
Figure 5. Changes in ice cover duration and the sum of negative air temperatures (from November to March) in two climate normal periods (shorter periods of Svy-Gun, Lėv-Ber, and Tat-Tre).
Figure 5. Changes in ice cover duration and the sum of negative air temperatures (from November to March) in two climate normal periods (shorter periods of Svy-Gun, Lėv-Ber, and Tat-Tre).
Sustainability 16 00725 g005
Table 1. Main hydro-morphometric characteristics of the selected rivers in 1961–2020.
Table 1. Main hydro-morphometric characteristics of the selected rivers in 1961–2020.
NoRiverWGSAbbreviationCharacteristics at WGS
Distance from River Mouth, kmBasin Area (A), km2Specific Discharge
(q), L/s km2
1.NemunasDruskininkaiNem-Dru447.737,4005.4
2.NemunasNemajūnaiNem-Nem336.742,9005.8
3.NemunasSmalininkaiNem-Sma111.081,1006.1
4.VerknėVerbyliškėsVer-Ver13.7694.37.2
5.NerisVilniusNer-Vil163.815,2006.4
6.NerisJonavaNer-Jon39.024,5006.6
7.ŽeimenaPabradėŽei-Pab17.526007.8
8.ŠventojiUkmergėŠve-Ukm39.853807.2
9.NevėžisPanevėžysNev-Pan133.011306.2
10.ŠušvėŠiaulėnaiŠuš-Šia108.6162.47.2
11.ŠušvėJosvainiaiŠuš-Jos14.210805.1
12.DubysaLyduvėnaiDub-Lyd75.010707.8
13.JūraTauragėJūr-Tau42.8166013.2
14.AkmenaPaakmenysAkm-Paa29.4314.013.6
15.ŠešuvisSkirgailaiŠeš-Ski5.218808.0
16.MinijaKartenaMin-Kar91.1122013.5
17.SvylaGuntauninkaiSvy-Gun9.7148.06.0 *
18.Nemunėlis TabokinėNem-Tab68.927407.1
19.MūšaUstukiaiMūš-Ust56.122804.5
20.Lėvuo BernatoniaiLėv-Ber47.211406.0 **
21.TatulaTrečionysTat-Tre5.5404.47.0 ***
22.VentaPapilėVent-Pap252.115606.2
23.VentaLeckavaVent-Lec186.240207.3
24.BartuvaSkuodasBar-Sku48.7616.712.1
* In the period of 1963–2020. ** In the period of 1967–2020. *** In the periods of 1961–1987 and 1993–2020.
Table 2. Results of homogeneity analysis: the breaking points in the series of average annual air temperatures and the series of the sum of negative air temperatures.
Table 2. Results of homogeneity analysis: the breaking points in the series of average annual air temperatures and the series of the sum of negative air temperatures.
ParameterPettittSNHT TestBuishand
pYearpYearpYear
Annual air temperature<0.00011988<0.00011988<0.00011988
Sum of negative air temperatures0.00319860.00419860.0021986
Table 3. Relationship between freeze-up dates and teleconnection patterns in the selected months.
Table 3. Relationship between freeze-up dates and teleconnection patterns in the selected months.
Teleconnection
Patterns		Jul–SepAug–OctSep–NovOct–DecNov–Jan
NAO0.03−0.070.050.110.24
AO0.160.070.110.090.18
EA0.230.260.260.200.13
SCAND−0.11−0.10−0.100.02−0.01
POL–EUR0.06−0.01−0.13−0.14−0.11
Table 4. Relationship between ice cover duration and teleconnection patterns in the selected months.
Table 4. Relationship between ice cover duration and teleconnection patterns in the selected months.
Teleconnection PatternsNov–FebNov–MarNov–Apr
NAO−0.58−0.59−0.57
AO−0.45−0.48−0.45
EA−0.39−0.37−0.35
SCAND0.190.250.31
POL–EUR0.260.310.30
Table 5. Relationship between ice break-up dates and teleconnection patterns in the selected months.
Table 5. Relationship between ice break-up dates and teleconnection patterns in the selected months.
Teleconnection PatternsDec–FebJan–MarFeb–Apr
NAO−0.44−0.54−0.40
AO−0.40−0.44−0.34
EA−0.40−0.34−0.21
SCAND0.190.260.30
POL–EUR0.140.190.24
Table 6. The Wilcoxon rank sum test results for TPs signal detection in the freeze-up dates in the selected months.
Table 6. The Wilcoxon rank sum test results for TPs signal detection in the freeze-up dates in the selected months.
River-WGSNAOAOEASCANDPOL–EUR
Oct–DecNov–JanDec–FebOct–DecNov–JanDec–FebJul–SepAug–OctSep–NovJul–SepAug–OctSep–NovSep–NovOct–DecNov–Jan
Ver-Ver 0.1 0.05
Nev-Pan 0.01 0.01 0.010.010.05
Šuš-Šia 0.1 0.05 0.010.010.01 0.1
Šuš-Jos 0.010.010.010.1 0.05 0.050.05
Šeš-Ski 0.05 0.05
Ven-Pap 0.010.050.01 0.1
Mūš-Ust0.10.010.050.10.010.10.010.010.01 0.10.1
Lėv-Ber0.10.050.01 0.010.10.050.050.1
Svy-Gun 0.10.01 0.05 0.1
Table 7. The Wilcoxon rank sum test results for TPs signal detection in the ice cover duration data.
Table 7. The Wilcoxon rank sum test results for TPs signal detection in the ice cover duration data.
River-WGS NAOAOEASCANDPOL–EUR
Nov–FebNov–MarNov–AprNov–FebNov–MarNov–AprNov–FebNov–MarNov–AprNov–FebNov–MarNov–AprNov–FebNov–MarNov–Apr
Ver-Ver0.10.05 0.1 0.10.05 0.1
Nev-Pan0.010.010.010.010.010.050.050.10.01 0.050.010.050.050.05
Šuš-Šia0.010.010.010.050.10.050.010.010.01 0.050.1 0.10.05
Šuš-Jos0.010.010.010.010.050.10.010.050.01 0.050.010.050.01
Šeš-Ski0.010.010.010.050.050.10.010.010.01 0.050.10.05
Ven-Pap0.010.010.010.010.010.010.050.10.01 0.010.010.010.10.1
Mūš-Ust0.010.010.010.050.050.050.010.010.010.10.05 0.050.010.05
Lėv-Ber0.010.010.010.010.010.010.050.10.050.050.010.1
Svy-Gun0.010.010.010.050.010.010.050.050.05 0.01 0.10.05
Table 8. The Wilcoxon rank sum test results for TP signal detection in the break-up dates.
Table 8. The Wilcoxon rank sum test results for TP signal detection in the break-up dates.
River-WGSNAOAOEASCANDPOL–EUR
Dec–FebJan–MarFeb–AprDec–FebJan–MarFeb–AprDec–FebJan–MarFeb–AprDec–FebJan–MarFeb–AprDec–FebJan–MarFeb–Apr
Ver-Ver0.010.010.050.10.050.10.05
Nev-Pan0.010.010.010.10.05 0.010.050.05 0.01
Šuš-Šia0.010.010.010.050.010.050.010.010.01 0.010.01 0.10.01
Šuš-Jos0.010.010.010.050.050.10.05 0.10.1 0.1
Šeš-Ski0.010.010.050.050.05 0.010.05
Ven-Pap0.010.010.010.050.010.10.050.01 0.05 0.1
Mūš-Ust0.10.010.010.10.010.010.010.010.1 0.050.01
Lėv-Ber0.10.010.010.10.050.010.010.010.01 0.010.01
Svy-Gun0.050.010.010.050.010.010.010.050.05 0.010.01 0.10.1
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Šarauskienė, D.; Jakimavičius, D.; Jurgelėnaitė, A.; Kriaučiūnienė, J. Warming Climate-Induced Changes in Lithuanian River Ice Phenology. Sustainability 2024, 16, 725. https://doi.org/10.3390/su16020725

AMA Style

Šarauskienė D, Jakimavičius D, Jurgelėnaitė A, Kriaučiūnienė J. Warming Climate-Induced Changes in Lithuanian River Ice Phenology. Sustainability. 2024; 16(2):725. https://doi.org/10.3390/su16020725

Chicago/Turabian Style

Šarauskienė, Diana, Darius Jakimavičius, Aldona Jurgelėnaitė, and Jūratė Kriaučiūnienė. 2024. "Warming Climate-Induced Changes in Lithuanian River Ice Phenology" Sustainability 16, no. 2: 725. https://doi.org/10.3390/su16020725

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop