







































Georgian Geographical journal 2022, Vol.2 (1) 34-40 

 

Impact of Global Warming on Zero Degree Isotherm of Near 

Surface Temperature 
Kukuri Tavartkiladze1*, Nana Bolashvili1 

Abstract 

The evaluation of the modern global warming impact on the zero degree isotherm crossing dates of near-surface 

temperature is unequivocally determined by the length of the warm (cold) season in a given region. The study was 

conducted using perennial, near-surface temperature field data from ten observation points in a complex, highland 

region. The correlation changes of the dates of the crossing of the zero degree isotherm of temperature concerning 

the vertical and horizontal displacement is studied. The statistical structure of the multi-year change in the dates 

is established. It is accepted that the increase in the warm season caused by global warming mainly occurs in the 

first half of the year when the zero-degree isotherm crosses from negative to positive. There are also rare cases 

when the warm season, for a long time, experiences a decrease in the opposite. 

Keywords: Surface temperature, global warming, warm (cold) season 

Introduction 

For more than a century, the Earth's energy balance has been disturbed - it receives more energy than 

it emits. The result of the gradual increase of energy potential is global warming. The main factors of 

global warming - solar energy, gaseous composition of the atmosphere, physical (optical) properties of 

the underlying surface - cause the increase of the Earth energy potential as the isolated body, can reveal 

completely different effects by regions [1-3]. There are regions where centres of cooling have emerged 

in the face of global warming. Such a region can be found in the small area of Georgia. In parallel with 

global warming, the coastal temperature in western Georgia has been gradually decreasing for decades 

[4, 5]. The process that creates the near-surface temperature field regime depends on the number of 

instantaneous and long-acting factors. 

The study of regime structure change, especially in such complex relief conditions as in Georgia, is 

the constant object of research. It is essential to assess the impact of modern warming on extreme 

deviations in near-surface temperature variations. Interestingly, for the given region, the extreme 

deviation change by time may be opposing, depending on what we account for the extreme deviation 

[6].  

Methods and Materials 

This paper aims to study the effects of modern global warming on the crossing dates of zero degree 

isotherm by the near-surface temperature. Such crossings occur at least twice at about one kilometre 

above sea level in the first and second half of the year. These dates are significant for the region as they 

determine the length of a given year's warm and cold periods. Particular attention should also be paid 

to the seasonal temperature change [7, 8, 9]. The critical indicator of seasonal change is crossing the 

zero degree isotherm by near-surface temperature.  

Systematic observational data of 10 meteorological stations in the 1070 -3659 km range in Georgia 

for 1906-2009 are used to solve the given task. 

As it is known, the average temperature can cross the zero degree isotherm not twice but several 

times per year. The reason for this is that the random factors result from short-term weather changes. 

Therefore, to determine the regime structure of the zero degree isotherm crossing date, it is not the 

average daily temperature directly used, but the inter annual temperature change constructed by the 

average monthly temperature. At the beginning of each year and the end of each observation point, the 

temperature transitions from negative to positive and from positive to negative are determined for the 

corresponding months. The zero degree isotherm crossing date is calculated by linear approximation 

and analytically. The range of calculated dates covers almost the entire year. This is only one point of 

observation (highland Kazbegi), which is located three and a half kilometres above sea level, and there 

were years when the crossing of the zero degree isotherm took place even in the summer months. 

 
1 Vakhushti Bagrationi Institute of Geography, TSU, Tbilisi, Georgia 

* Corresponding author: kuktav@mail.ru 

 



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

Results and Discussion 

The main effect on the transition date from a negative value to a positive one is the temperature 

regime of the first half of the year. In the second half, on the contrary, the date of the transition from 

positive to negative is determined by the regime structure of the second half. The characteristics of the 

temperature field change in the first and second half of the year can be assessed by Fig. 1. The figure 

shows the temperature field change, based on 1906-2009 data for ten observation points. 

 
Figure 1. The normalized near surface temperature change of first (1) and second (2) half of the year by the linear and non-linear 

approximations and corresponding standard deviations (dashed lines) for Georgian highland 

In Fig.1, for the first and second half of the year, the normalized temperature changes based on all 

observation points data are presented, with linear and nonlinear approximations (as the month 

normalized parameter the monthly arithmetic mean has been taken of this observation point for each 

observation point, each year). The change direction (cooling or warming) and the magnitude is 

determined by the so-called "dynamic norm" [10], which can be used to assess the accuracy of the 

approximation. For example, the representation of temperature change in time by dynamic norm 

compared to the static norm reduces the standard error by about 5% in the case study (Fig. 1). Moreover, 

when the temperature goes from negative to positive, the error is reduced by 8%. 

The line equation determines the dynamic norm. When switching from negative to positive, it 

became the following expression: 

            T(n) = 0.245 + 0.00038 n,                                       (1) 

And the transition from positive to negative, e.g., in the second half of the year 

                  T(n) = 0.722 + 0.00014 n,                                 (2) 

      where T(n) is the mean temperature of 6 months, n- year  

From the equations, it is evident that the near-surface temperature change was of low intensity in the 

first and especially in the second half of the year of the highland system. In particular, according to (1) 

and (2), their values are + 0.038°C / 100 year and + 0.014°C/ 100 year. 

The nonlinear approximate curves given in Fig.1 are of seventh order polynomials. These 

polynomials give the least standard errors when describing temperature changes compared to other 

polynomials. However, it should be noted that in this case, they are only 1.5 - 2.0% more accurate than 

the linear approximation in determining the temperature change result. 

The mean, minimum and maximum intersection dates of the zero degree isotherm for the ten 

observation points of the highland zone of Georgia are given in Table 1. 



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

 
Figure 2. The dependence of warm season duration on place elevation 

As seen from the table, in the case of four observation points, the maximum date of crossing the zero 

degree isotherm is not fixed until the end of the year, which means that it moves on to the following 

year. 

To determine the relationship of the warm season duration of the year to the place elevation above 

sea level, the average multi-year crossing dates of the zero degree isotherm by temperature were 

determined. Obtained results are used in Fig. 2, which shows the mean date changes according to the 

place elevation.  

As can be seen from the drawing, in the highlands of Georgia, the so-called warm season, at about 

one mile above sea level, exceeds nine months. As the height increases, it decreases linearly and 

becomes less than three months at the three and a half kilometre elevation. Thus, above 1000 meters, 

as the place elevation increases with every 100 meters, the warm season decreases by 7-8 days. 

The obtained result is typical for a highland area covering several hundred square kilometres. It is 

obtained by combining multiannual data from 10 observation points. In such conditions, it is of interest 

to determine the relationship between observation points in the variations of the zero degree isotherm 

crossing dates. For this, we constructed the autocorrelation matrix with the annual dates of all 

observation points that determined the crossing of the zero degree isotherm for both the first and second 

half of the year. It is given in Figure 3. 

Table 1. The zero degree isotherm crossing dates by temperature for Georgian highland (1906-2009) 

 

 

 

N 

 

Observation 

post 

Elevation 

Zero degree isotherm crossing data (date, month) 

 

Transition from negative to positive 

temperature 

 

Transition from positive to negative 

temperature 

Avg. min max Avg. min max 

1 Pasanauri 1.07 27.02 30.01 21.03 10.12 9.11 - 

2 Dmanisi 1.256 24.02 18.01 24.03 21.12 12.11 - 

3 Abastumani 1.265 10.03 9.02 28.03 30.11 5.11 31.12 

4 Mestia 1.441 14.03 14.02 6.04 24.11 1.11 14.12 

5 Tsalka 1.457 12.03 12.02 1.04 3.12 4.11 - 

6 Bakhmaro 1.926 23.03 2.03 20.04 28.11 29.10 - 

7 Paravani 2.1 7.04 13.03 20.04 13.11 23.10 3.12 

8 Gudauri 2.194 6.04 3.03 23.04 13.11 11.10 22.12 

9 Jvari Pass. 2.395 2.06 3.04 18.05 26.10 1.10 22.11 

10 Kazbegi 3.659 14.06 10.05 6.07 17.09 28.08 14.10 



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

The observation points on the horizontal axes are placed according to the place height increase from 

sea level. In general, especially in the first half of the year, there is a gradual decrease in correlation 

with increasing height distance. Nevertheless, there are certain heights where this natural process is 

disrupted. As it can be seen from the drawing, such processes are not random. It is permanent for the 

period under consideration and may be caused by the constant exposure of a different landscape or near-

surface to its energy potential redistribution. 

Fig. 3 shows the vertical variations of the zero degree isotherm crossing date. To identify what these 

variations are in the case of horizontal displacement, the correlation coefficients were determined by 

considering the distance between the observation points. The obtained results are given in Fig.4. 

 

Observation points, sorted by heigh above sea level 

 
Figure 3. Correlation of zero degree isotherm crossing dates by height between observation points (1,2,3, ..., 10) in the first (1) and second 

(2) half of the year. 

This connection, as expected, will not be significantly changed. It varies approximately in the range 

of 0.4 to 0.6 correlation coefficients by amount. This is entirely natural, as the maximum distance 

between observation points does not exceed 250 km. 

As mentioned above (Fig. 1), the near-surface temperature change is occurring at very low intensity 

in the highland zone of Georgia. 

 
Figure 4. Correlation of zero degree isotherm crossing dates by direct distances between observation points in the first (1) and second (2) 

half of the year (dashed lines define standard errors). 

To determine what impact has the near-surface temperature changed over the date of the zero degree 

isotherm. In addition, warm-season duration and the impact of global warming on it were also 

determined for each observation point. The obtained results are given in Table 2. 

1      2      3      4      5      6      7      8      9    ` 1      2      3      4      5      6      7      8      9  

Observation points, sorted by heigh abuve sea level

0.0

0.2

0.4

0.6

0.8

1.0

C
o
rr

el
a
ti

o
n

  
c
o
e
ff

ic
ie

n
t

(1) (2)



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

Table 2. The impact of global warming on the change of  zero degree isotherm crossing date 

 

 

 

 

N  

 

 

Observation post 

Zero degree isotherm date 

change (day /100y) 
Warm season duration 

Transition 

from 

negative to 

positive 

temperature  

 

Transition from 

positive  to 

negative 

temperature 

Avg. (day) 
Change 

(day/100y) 

1 Pasanauri -9.7 5.9 280 16 

2 Dmanisi -14.2 15.4 307 29 

3 Abastumani -6.0 -1.3 269 5 

4 Mestia -3.5 -3.3 255 0 

5 Tsalka -6.6 6.4 269 13 

6 Bakhmaro -1.0 -2.8 246 -2 

7 Paravani -3.8 0.6 220 5 

8 Gudauri -1.0 -2.6 224 -2 

9 Jvari Pass. 
-7.1 1.7 161 

9 

 

10 Kazbegi 
-8.3 4.0 94 12 

 

As evident from the table, the changes of the emergence dates of the zero degree isotherm in the first 

and second half of the year are heterogeneous. Whereas the dates of the transition from negative to 

positive temperature, due to the global warming, recede in the case of all observation points, i.e. shifts 

in the direction of increase of the warm season duration, changing from positive to negative, in four to 

ten cases the opposite happens. Nevertheless, the duration of the warm-up season was not reduced for 

the two observation points (Abastumani, Mestia), as the duration was longer in the first half of the year 

than in the second half. However, in two observation points (Bakhmaro, Gudauri), the process of global 

warming was marked by a decreased tendency during the warm season. 

As mentioned, the change of the warm season duration by height is subject to a definite regulation. 

As for the change by time or the change that may be caused by global warming, it is not subject to any 

regularity. Indeed, warming generally prolongs the warm season, but it is challenging to substantiate 

the reasons for the two not-so-distant regions (Dmanisi, Abastumani) in one, the duration of the warm 

season has been increased in 100 years by the month, and in the other - by a few days. 

The crossing dates changes of the zero degree isotherm over the years do not differ significantly for 

the whole study area. Fig. 5 shows the change of one of them (Pasanauri) for the study period for both 

the first and second half of the year. It is selected from the point of view that the variations in the dates 

of the crossing of the zero degree isotherm of temperature, both in the first and second half of the year, 

are more diverse from the general picture than the others. 

Dates are marked on the vertical axes of the drawings. In both cases, their change range rarely 

exceeds 30 days. Linear and nonlinear approximations are used to determine the changing trend. The 

accuracy of the approximations is estimated using standard errors. 



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

 
Figure 5. The transition from negative to positive (1) and from positive to negative (2) date change by years of near surface temperature of 

Paravani using 1906-2009 year data and representation of those changes by linear (dashed) and nonlinear (curved) approximations 

The linear approximation of the first half of the year improves the accuracy by 7.51% compared to 

the arithmetic mean (i.e. when no change is foreseen at all). The nonlinear approximation was 

performed using a polynomial, the optimal expression of which was determined by the gradual increase 

of the polynomial order. Typically, the maximum accuracy of a polynomial was reached when the 

polynomial was of 7 or 8 order (all nonlinear approximations in the present paper are of the seventh 

order). In the example given, the nonlinear approximation improved the accuracy by 0.06% compared 

to the linear approximation. As for the case of the second half of the year, i.e. when the temperature 

crosses the zero degree isotherm from positive to negative, the accuracy of the determined values are 

the following: The linear approximation reduces the standard error by 2.35% compared to the arithmetic 

mean. Moreover, by nonlinear approximation, the standard linear approximation error is improved by 

only 0.14%.  

Dates are marked on the vertical axes of the drawings. In both cases, their change range rarely 

exceeds 30 days. Linear and nonlinear approximations are used to determine the changing trend. The 

accuracy of the approximations is estimated using standard errors. 

 
Figure 6. The crossing date change by years of near surface temperature and it representation by linear (dashed) and nonlinear (curved) 

approximations while temperature transition from negative to positive (1) and from positive to negative (2) 

The annual dates change of crossing the zero degrees isotherm by the near-surface temperature field 

over the years is represented by the normalized values of all observation points. The changes based on 

the combined data by years are given in Figure 6. First of all, the images show that the dispersion in the 

second half of the year is almost twice less than in the first half, i.e. the transition from a positive to a 

negative value is stable in a much shorter time. However, the impact of global warming on this process 

is not revealed at all. That means that the transition date from positive to negative temperature value 

does not change. The prolongation of the warm season occurs only during the transition from a negative 

value to a positive one. 

  



Tavartkiladze K. Georgian Geographical journal 2022, Vol.2 (1) 

Conclusion 

To assess the climate change impact on the duration of the warm season, the nonlinear approximation 

compared to the linear one gives almost no improved results. 

Competing interests 

The authors declare that they have no competing interests. 

Authors’ contribution 

All authors discussed the results and contributed to the final manuscript. K. T. conceived the idea; 

took the lead in writing the manuscript. N. B. provided critical feedback and helped shape the analysis 

and manuscript, worked out almost all of the technical details.  

References  

[1] Budiko M. Klimat v proshlom I budushem [Climate in the past and the future]. L., Gidrometeoizdat, 1980, 

352 p. (in Russian) 

[2] Kondratyev K. Globalnii klimat [Global climate], Nauka, 1992, 352 p. (in Russian) 

[3] Lovejoy T.E. and L. Hannah, eds., Climate Change and Biodiversity. Yale University Press, London. 2005, 

pp. 440  

[4] Amiranashvili A.G., Amiranashvili V.A., Gzirishvili T.G., Kharchilava J.F., Tavartkiladze K.A. Modern 

Climate Change in Georgia. Radioactively Active Small Atmospheric Admixtures. Tbilisi, 2005 (128).  

[5] Tavartkiladze K., Begalishvili N., Tsintsadze T., Kikava A. Influence of Global Warming on the Near 

Surface Air Temperature Field in Georgia. Bulletin of the Georgian National Academy of Sciences, v.6, N 

3, 2012 (55-60).  

[6] Sherstyukov B.G. Regionalnie I sezonnie zakonomernosti izmenenii sovremennogo klimata [Regional and 

seasonal change regulations of the modern climate]. "VNIIGMI-MCD", 2008, 246 p. (in Russian) 

[7] Loginov V.F. Sezonnie osobennosti mnogoletnix izmenenii globalnogo klimata I ikh prichini [Seasonal 

features of long-term changes of the global climate and their causes]. Ukrainian Geographical Journal, No. 

2, 2013 (26-239). (in Russian) 

[8] Tavartkiladze K., Begalishvili N., Tsintsadze T., Kikava A. Influence of Global Warming on the Near-

Surface Air Temperature Field in Georgia. Bulletin of the Georgian National Academy of Sciences, v.6, N 

3, 2012 (55-60). 

[9] Tavartkiladze K., Kikava A. (2021). Annual Amplitude Regime of Ground Surface Temperature Field in 

Georgia. Georgian Geographical Journal, 1(1).  https: //journals.4science.ge/index.php/GGJ/article/view/260 

[10] Tavartkiladze K. Ukiduresi gadakhrebi mitsispira temperaturuli velis variatsiebshi da misi kavshiri havis 

tsvlilebastan (tbilisis magalitze) [The extreme deviations of near surface temperature variations and its 

connection with climate change]. Transactions of the Institute of Geography v.3, 2011, pp.39-45. (in 

Georgian). 

 

 

 

https://journals.4science.ge/index.php/GGJ/article/view/260
https://journals.4science.ge/index.php/GGJ/article/view/260

