CC BY
21
Kholmatjanov Bakhtiyar Makahamatjanovich, E-mail: [email protected] Petrov Yuriy Vasilevich, E-mail: [email protected] Jaloliddinov Bunyodjon Makhmudjon ogli, National University of Uzbekistan named after Mirzo Ulugbek
E-mail: [email protected]
TRENDS AND VARIABILITY OF AIR TEMPERATURE AND PRECIPITATION FROM 1961 TO 2016 IN THE KYZYLKUM DESERT
Abstract: Comprehensive analysis of variability of the temperature and precipitation regime in Kyzylkum desert in the course of two last climatic periods made it possible to reveal peculiar features of these variations for separate half years and a year as a whole, and to evaluate their significance, as well. It is shown that the change of climatic conditions within homogeneous landscape can occur in asynchronous way. It is foreseen that increase of the air temperature variability in the warm half-year can be considered as a feature ofbifurcation period — i.e., change from warming to cooling in the nearest feature.
Keywords: climatic period, warming, air temperature, precipitation, trend, Kyzylkum desert, bifurcation.
Introduction
In the Fifth Assessment Report of IPCC it is emphasized that «Warming of the climate system is unequivocal, and since the 1950s, many of the observed changes are unprecedented over decades to millennia. The atmosphere and ocean have warmed, the amounts of snow and ice have diminished, and sea level has risen» [2]. Spatial-and-tem-poral distribution of temperature and precipitation, their tendencies and variability related to climate change caused by the global warming and anthropogenic activities have raised significant attention, especially in arid and semi-arid
regions where the climatic indices are very important environment factors. In the aggregate, these factors can have substantial impact on ecological systems and on life activity of population in regional scale, in a whole. That is why, the studies of the temperature and precipitation regime and of their tendencies and variability in one of the arid region of Uzbekistan have rather significant scientific and practical importance.
Study area. From the point of view of distinguishing homogeneous climatic zones, Uzbekistan is subdivided to 14 regions considering their physical-and-geographic conditions.
Figure 1. Schematic map of the study area and location of meteorological stations
Variations of meteorological conditions in these regions are regarded as homogeneous. That is why, for each of these regions the most representative meteorological stations are determined as the data ofthese stations observations characterize the regime of meteorological quantities for the whole region. The study area covers non-irrigated territory of Kyzylkum desert of Uzbekistan. According to this zoning Kyzylkum desert includes 4 climatic regions, 3 of which are referred to as non-irrigated: Northern, Central and Southern Kyzylkum (Figure 1).
Materials and methods. Akbaytal (Northern Kyzylkum), Tamdi (Central Kyzylkum) and Ayakagitma (Southern Kyzylkum) are representative meteorological stations for the investigated climatic regions. Investigations of the regimes of the air temperature and precipitation and of their tendencies and variability are performed on the base of the data of these
stations observations in the period of 1961-2016. For the processing of data series of observations of the air temperature and precipitation the methods for statistic and comparative analyses are applied. Analysis of the trend changes and of variability of air temperature and precipitation is made on the base of non-parametric statistic test of Mann-Kendall [1, 3-5; 7; 8]:
Q = ■
a
Iii
i(n-2)(2n + 5)-^x,( -2)(2x, + 5)
, (1)
where n is record length, a is standard deviation, x. is the number of identical data values (for each value).
The confidence levels of the trend variations according to the values of non-parametric statistical test of Mann-Kendall are given in (Table 1).
Table 1. Confidence levels CL and error probabilities a for the Mann-Kendall nonparametric statistical test
Q CL a Q CL a
> 1 > 68.3% < 0.317 > 1.282 > 80% < 0.2
> 1.5 > 86.6% < 0.134 > 1.645 > 90% < 0.1
> 2 > 95.4% < 0.046 > 1.960 > 95% < 0.05
> 3 > 99.7% < 0.003 > 2.576 > 99% < 0.01
> 4 > 99.999% < 0.00001 > 3.290 > 99.9% < 0.001
Results and discussion. For the purposes of statistical 1990 and 1991-2016. It is known that the first time period is and comparative analyses the time series of the air tempera- taken as the base climate period and is used for assessment of ture and precipitation were divided into two periods - 1961- the scale of climate change.
Table 2. - Mean monthly, half-year and yearly values of the air temperature and precipitation during 1961-1990 (1) and 1991-2016 (2); Air temperature (°C)
Station Period Month CHY WHY Year
I II III IV V VI VII VIII IX X XI XII
Akbaytal 1 -6.0 -4.2 3.9 13.9 21.2 27.0 29.7 26.7 20.0 10.7 3.4 -2.8 1.4 22.6 12.0
2 -5.0 -2.5 5.6 14.7 21.9 27.8 29.6 27.8 20.7 12.2 3.4 -2.8 2.2 23.3 12.8
Tamdi 1 -2.5 -0.5 6.7 15.6 22.5 28.5 31.0 28.3 21.7 12.9 6.0 0.4 4.3 24.2 14.2
2 -1.3 0.9 8.2 16.3 23.0 29.1 30.9 29.2 22.3 14.1 6.0 0.2 5.1 24.8 14.9
Ayakagitma 1 -1.7 0.8 7.6 16.0 23.0 28.8 31.4 28.8 22.0 13.1 6.0 0.6 4.9 24.5 14.7
2 -1.0 1.3 8.3 16.0 23.1 29.2 31.1 29.3 22.2 13.6 5.8 0.4 5.1 24.7 14.9
Precipitation (mm)
Station Period Month CHY WHY Year
I II III IV V VI VII VIII IX X XI XII
Akbaytal 1 9.6 11.9 17.3 21.9 11.8 3.5 3.8 1.7 2.2 6.8 11.4 16.7 88.8 29.8 118.6
2 12.8 12.8 17.9 18.5 19.8 7.8 2.0 1.9 0.7 4.8 14.5 8.6 85.1 37.0 122.0
Tamdi 1 12.7 16.4 21.5 23.0 13.1 2.6 2.9 0.5 1.2 5.4 10.0 16.6 100.2 25.7 125.9
2 10.3 15.6 16.9 14.9 12.9 4.5 1.9 2.7 0.4 4.3 13.7 13.1 84.5 26.7 111.2
Ayakagitma 1 14.2 15.2 27.8 26.7 13.3 1.8 2.8 0.1 1.0 7.4 11.4 18.2 113.5 26.4 139.9
2 15.6 22.1 24.7 16.3 12.3 3.1 2.1 1.4 0.5 5.4 17.8 13.4 109.8 24.8 134.5
Note: CHY - cold half-year, WHY - warm half-year.
14,0
13,5
a. £
aj
f 11,5 a
c 11,0 c
^ 10,5
10,0
~ 13,0 ■ 1
I 12,5 i _
^ «A
¡1 12,0 VI \
G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)G)
<N<N<N<N<N<N<N<N
a)
-Annual temperature 1961-1990
-Average annual temperatu re 1961-1990
--Annual temperature linear trend in 1991-2016
-Annual temperature 1991-2016
--Annual temperatu re linear trend in 1961-1990
17,0
16,5
U 16,0 0
i 15,5 S 15,0 14,5 14,0 13,5 13,0 12,5
<1>
.
4
11,5
TA1= 0,0071t+0,2615
-4----------j- --------------J —a"A—A ■Pf \ \A-j--r]
f"l T\ A/ V / \ / f-----------------------1-
" TA2= 0,0362t-57,654
CO Ifi N 0) CD CD CD CD CD
CO Ifi N 0)
cococococooooo)
0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)
CO in
o o o o
CNCNCNCNCNCNCNCN
b)
-An nual temperatu re 1961-1990
-Average annual temperatu re 1961-1990
--Annual temperature linear trend in 1991-2016
-Annual temperature 1991-2016
--Annual temperature linear trend in 1961-1990
16,5
16,0
0
», 15,5
= 15,0 ■----
<0
(U 14,5 .
| 14,0
+J
n 13,5
3
c 13,0 <
12,5
12,0
oooooooooooooooooooo
<N<N<N<N<N<N<N<N
c)
-Annual temperature 1961-1990
-Average annual temperature 1961-1990
--Annual temperatu re linear trend 1990-2016
-Annual temperature 1991-2016
--Annual temperature linear trend 1961-1990
Figure 2. Annual changes of temperature from 1961 to 2016 in the Kyzylkum desert. (a) Akbaytal; (b) Tamdi; (c) Ayakagitma
The comparison of the mean monthly values of the air temperature in the above mentioned periods shows that their
change within the studied climatic regions was uneven (Table 2). In all climatic regions the highest increase of the mean
monthly temperature was observed from January to March and in October. With this, in the regions of the Northern and Central Kyzylkum the value of accretion was 1 °C and more. Finally, the records show that that considering half-years and annual values, the increase of the mean annual air temperature was 0.8, 0.7 and 0.8 °C in the Northern Kyzylkum, 0.8, 0.6 and 0.7 °C in the Central Kyzylkum and 0.2, 0.2 and 0.2 °C in the Southern Kyzylkum, respectively. Uneven change is also observed in the value of precipitation amount. In the Northern Kyzylkum the annual precipitation amount increased (+3.4 mm), while in two other regions it decreased (-14.7 and -5.4 mm, respectively). As it shown in Table 2, the decrease of precipitation amount took place during the cold half-year in all studied climatic regions. Insignificant increase of precipitation amount in the Northern Kyzylkum was determined with the increase of this value in the warm half year.
Thus, the increase of the mean annual air temperature was observed, mainly, during the months of the cold half-year. For the analysis of this circumstance and for determination of tendencies of the air temperature variations the graphs of the interannual variation of the mean half-year and annual values of the air temperature (Figure 2) and annual precipitation amount were constructed (Figure 3).
Figure 2 shows that during the base (1961-1990) period in the regions of Northern (a) and Central Kyzylkum
(b) insignificant decrease tendency was observed, while in the Southern Kyzylkum (c) the decrease of the mean annual temperature was observed. Currently, in these climatic regions the constant increase of the air temperature is observed. On the base of derived linear regression equations and using non-parametric statistic test of Mann-Kendall the characteristics of tendencies of variations of the mean half-year and annual air temperature values (Table 3) and of annual precipitation amounts (Table 4) were obtained. As the data ofTable 3 show, during 1961-1990 the confidence levels of the trend variations are rather low (0.0 and <68.3%) with the exclusion of the warm half-year for the Northern Kyzylkum (>86.6%). During 1991-2016 the statistic test the value of Qfor all climatic regions is 1.00 and higher during the warm half-year and regarding the annual values. Respectively, the confidence levels of the trend variations, especially during the warm half-year for the Central and Southern Kyzylkum are >80.0 and >90.0%, respectively. It is just this circumstance, that the annual confidence levels of the trend variations of temperature in all three climatic regions are within the range of >68.3 — >80.0%.
Thus, though the increase of the mean annual air temperature was determined by the increase during the cold half-year, the rate of the increase of the mean annual air temperature in the warm half-year were more significant in the occurred variations of the mean annual air temperature.
Table 3. Characteristics of the tendencies of the variations of air temperature in Kyzylkum desert in 1961-1990
Station Season Trend equation Trend accretion, AT (°C) ff Q CL (%)
Akbaytal Annual TA = 0.0159t - 19.525 0.5 0.84 0.55 < 68.3
Cold half-year TC = 0.0008t - 0.1368 0.0 1.51 0.02 0.0
Warm half-year TW, = 0.0311t - 38.914 0.9 0.59 1.53 > 86.6
Tamdi Annual T4, = 0.0071t + 0.2615 0.2 0.82 0.25 < 68.3
Cold half-year TC = 0.0012t + 1.8892 0.0 1.46 0.00 0.0
Warm half-year TW, = 0.0129t - 1.3662 0.4 0.50 0.74 < 68.3
Ayakagitma Annual T4, = -0.0204t + 55.05 -0.6 0.87 0.68 < 68.3
Cold half-year TC = -0.0315t + 67.111 -0.9 1.56 0.59 < 68.3
Warm half-year TW = -0.0094t + 43 -0.3 0.54 0.50 < 68.3
1991-2016
Station Season Trend equation Trend accretion, AT (°C) ff Q CL (%)
1 2 3 4 5 6 7
Akbaytal Annual TA2 = 0.0381t - 63.584 1.0 0.89 1.07 > 68.3
Cold half-year TC = 0.0321t - 61.959 0.8 1.63 0.46 < 68.3
Warm half-year TW2 = 0.0442t - 65.208 0.8 0.80 1.00 < 68.3
Tamdi Annual T42 = 0.0362t - 57.654 0.9 0.74 1.23 > 68.3
Cold half-year TC = 0.0286t - 52.292 0.7 1.33 0.54 < 68.3
Warm half-year TW2 = 0.0438t - 63.016 1.1 0.80 1.37 > 80.0
1 2 3 4 5 6 7
Ayakagitma Annual TA2 = 0.0417t - 68.61 1.0 0.70 1.49 >80.0
Cold half-year TC2 = 0.0335t - 61.915 0.8 1.24 0.67 <68.3
Warm half-year TW = 0.05t - 75.305 1.3 0.68 1.85 >90.0
300,0 250,0 200,0 150,0 100,0 50,0 ■
CD CD CD CD CD.........................
0)0)0)0)0)0)0)b)0)0)d)d)did)cS^didi^di
a)
Annual precipitation 1961-1990 Average annual precipitation 1961-1990 Annual precipitation lineartren in 1991-2016
-Annual precipitation 1991-2016
--Annual precipitationlineartrend in 1961-1990
300,0
250,0 ■
c
0 200,0
£ 150,0
u
2!
5 100,0 0 3
c
c <
50,0
0,0
RA1 = 0,8558t - 1567,7 RA2 = -1,264t + 2641
A i I i — V—"Ar
A, hi —A——T\ —/
WAV/-------J^N—V..J..L.. —Yl "
CO Ifi N 0)
CO Ifi N 0)
CDCDCDCDCDNNNNNCOCOCOCOCOfflfflfflfflO)- _ _ _ 0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0000
CM CM CM CM
° S °
CM CM CM
b)
—Annual precipitation 1961-1990 —Average annual precipitation 1961-1990 - —Annual precipitation lineartrend in 1991-2016
Annual precipitation 1991-2016
Annual precipitation lineartrend in 1961-1990
300,0
250,0
c
0 200,0
u 2!
5 100,0 ra
3
c
c <
50,0
0,0
RA1 = 1,5721t-2971,8 RA2 = -0,108t + 350,9
V ri A
—\—T \ / \ /\ A A
\-N"\ 7\ r \ / S/ ^y*
XI
lfiN0)T-c01fiN0)T-c01fiN0)T-c01fiN0)T-c01fiN0) CDCDCDNNNNNCOCOCOCOCOfflfflfflfflfflOgggg 0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)
c)
Annual precipitation 1961-1990 Average annual precipitation 1961-1990 Annual precipitation lineartrend in 1991-2016
-Annual precipitation 1991-2016
--Annual precipitation lineartrend in 1961-1990
Figure 3. Annual changes of precipitation from 1961 to 2016 in the Kyzylkum desert. (a) Akbaital; (b) Tamdy; (c) Ayakagitma
The intrannual variations of precipitation amount in the studied climatic regions were of different character. The Northern Kyzylkym region was characterized with insignificant decrease of precipitation amount in the period of1961-1990 and with precipitation increase in the period of1991-2016 (Figure 3a). The Central and Southern Kyzylkum regions were characterized with the inverse tendency (Figure 3 b, c).
Characteristics of the tendencies of precipitation variations both in the period of 1961-1990, and in the period of 1991-2016 are of rather low confidence levels (Table 4). With
Conclusions. Analysis of the change of the regime of the air temperature and atmospheric precipitation in Kyzylkum desert during two last climatic periods gives the ground to make the following conclusions. First, in the first climatic period the insignificant warming was observed only in the northern and central parts of the region and only in the warm half-year. In the southern part of the desert the whole-year decrease of the air temperature was observed. Annual precipitation amount in the northern part of the desert also changed in asynchronous way over the whole territory. In the first period insignificant decrease of precipitation was observed in the northern part of the desert, while significant increase of
this, clearly expressed negative gradient of the trend accretion from the north to the south is traced. The reason of this circumstance is that during the movement of moisture laden air masses over the territory of Kyzylkum desert from the north to the south the warming of the air takes place from the underlying surface. In the result of this the air masses are not in saturation conditions more. As it was shown above, during the last decades the process of constant warming takes place, which undoubtedly, intensifies the negative gradient of precipitation from the north to the south.
precipitation was observed over its other parts. During the second period the inverse situation was observed: significant precipitation increase in the northern part and its decrease in the other parts of the desert. Thus, the proceeding of climatic changes, even within comparatively homogeneous landscape is rather complicated both temporally, and spatially, and also regarding the values of meteorological parameters.
Second, the increase of V by the air temperature In the warm half-year on the whole territory can be regarded as the feature of bifurcation period [6], for which the deepening of the process instability is typical. In this cased the change from the warming to cooling in the nearest future is meant.
Table 4. Characteristics of the tendencies of the variations of precipitation in Kyzylkum desert
Station Period Trend equation Trend accretion, AR (mm) Q CL (%)
Akbaital 1961-1990 RA1 = -0.2912t + 688.28 -8.4 38.30 0.22 < 68.3
1991-2016 RA2 = 1.761t - 3405.1 44.0 57.30 0.77 < 68.3
Tamdy 1961-1990 RA1 = 0.8558t - 1567.7 24.8 39.66 0.63 < 68.3
1991-2016 RA2 = -1.264t + 2641 -31.6 39.18 0.81 < 68.3
Ayakagitma 1961-1990 RA1 = 1.5721t - 2971.8 45.6 45.41 1.00 < 68.3
1991-2016 RA2 = -0.108t + 350.9 -2.7 42.70 0.06 0.0
References:
1. Chen Y., Deng H., Li B. et al. Abrupt change of temperature and precipitation extremes in the arid region of Northwest China [J]. Quat. Int. 336, 2014.- P. 35-43.
2. IPCC, 2014: Climate Change 2014: Synthesis Report. Contribution ofWorking Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R. K. Pachauri and L. A. Meyer (eds.)]. IPCC, Geneva, Switzerland, - 151 p.
3. Kendall M. G. Rank Correlation Methods. Griffin, - London. 1975.
4. Mann H. B. Nonparametric tests against trend. Econometrica 13, 1945.- P. 124-259.
5. Progress report to CCI on statistical methods. WMO-TD.- No. 834.- 116 p.
6. Thompson J. M.T., Sieber J. Predicting climate tipping as a noisy bifurcation: A review. Int. J. Bif. Chaos. 2011.- Vol. 21,-No. 2.- 399 p.
7. Yao J., Chen Y., Yang Q. Spatial and temporal variability of water vapor pressure in the arid region of northwest China, during 1961-2011 [J]. Theor. Appl. Climatol. 2015.- P. 1-9.
8. Yue S., Pilon P., Cavadias G. Power of the Mann-Kendall and Spearman's tests for detecting monotonic trends in hydrologi-cal series. Journal of Hydrology, 259: 2002.- P. 254-271.
