4.1. Air Temperature Analysis in the Varzob River Basin
The Varzob River Basin’s various regions’ monthly air temperatures are shown in
Figure 2. The Anzob station’s territory is in an area with a climate that is only moderately humid, with mild winters and relatively warm summers. The ambient temperature is −1.8 °C on a yearly average. The time without frost typically lasts 83 days. January is the coldest month with an average temperature of (−12.1 °C). The average minimum air temperature is −15.8 °C in January, however, severe cold air masses can cause it to plunge as low as −24–−36 °C. At the same time, throughout the winter, it can warm up to −5 °C during the day on some days. July is the hottest month (+9.7 °C). The maximum temperature during the hottest months is +24 °C, with the air warming up to +20 °C during the day. At the same time, if the minimum temperature is typically +5–−6 °C in the summer, it can occasionally fall to −4–−5 °C at night in the coldest years. In the summer, the absolute minimum temperature was −10 °C.
The Maykhura and Ziddi rivers meet at the Maykhura station area. The valley close to the station has rather steep slopes and is relatively narrow (200–300 m). The station is 1921 m a.s.l. The station’s territory is in an area with a climate that is moderately humid, with mild winters and pleasant summers. The first frosts appear in October, and the air-frost period lasts until mid-May. January is the coldest month of the year (average temperature of −8.6 °C). The severe cold air masses can cause the average minimum air temperature to plunge as low as −20–−27 °C. At the same time, during the daytime in the winter, the air temperature might drop to 0–−7 °C below zero on some days. July is the hottest month of the year (average temperature of +18.6 °C). The air warms up to +25–28 °C during the warmer months, with +35 °C being the absolute maximum. At the same time, if the minimum temperature is typically +7–−8 °C in the summer, it may occasionally fall to +3 °C at night in the coldest years. We discovered that the warmest month of the year is thought to be July, and the coldest month of the year is thought to be January by keeping track of the meteorological data (temperature) during the period 1960–1990.
The winter period in the middle part of the river basin is not very severe and the temperature background practically corresponds to the seasons. The calendar months of winter, namely December, January, and February, show relatively negative values. With the onset of spring, the temperature consistently shows a positive temperature.
Hushyori station’s region is in an area with a climate that is moderately humid, with pleasant summers and mild winters. The average yearly temperature is +11.2 °C. The time without frost is typically 225 days long. January is the coldest month. The minimum air temperature is typically −5.2 °C, although it can fall as low as −20–−27 °C when huge cold air masses arrive. At the same time, it can get warmer during midday in the winter on some days, reaching +5–14 °C heat. July is the hottest month with +24.2 °C (average temperature). The maximum temperature in the hottest months is +39 °C, with the air warming up to +25–30 °C during the day. In contrast, if the minimum temperature is typically +13–−16 °C in the summer, it may occasionally fall to + 3 °C at night in the coldest years.
By monitoring the meteorological data of the two periods (1960–1990 and 1991–2018), we found out that July is the warmest month and January is the coldest month of the year. The winter period in the lower reaches of the river is not very severe and the two remaining months of winter are December and February. The spring–autumn period practically does not cause sharp jumps in temperature.
Assessing the temperature values by years at Anzob station in the upstream region, we can say that the predominance of sub-zero temperatures is clearly visible on the graphs and the relative minimum is below zero (
Figure 3). General temperature monitoring for the period 1990–2018 showed that the trend line is in the mode of stable sub-zero temperatures. Additionally, small temperature spikes in 1998 and 2006, when the temperature jumped out of the zero mark, showed us relative anomalous years.
During the period under consideration, several years with a relatively hot weather were observed at Maykhura station: 1966, 1970, 1971, 1973, 1977, 1979, 1980, 1987, 1988, 1990. Additionally, in contrast to them, rather cold year weather periods were observed: 1964, 1967, 1969, 1972, 1974, 1975, 1976, 1981, 1982, 1984, 1989. During the period 1990–2018, in several years, relatively hot weather was observed at Maykhura station including 1997, 2002, 2006, 2007, 2008, 2010, 2014, 2015, 2016, and 2017. On the contrary, the cold weather periods were observed for the following years: 1991, 1992, 1993, 1995, 1996, 1998, 2004, 2005, 2009 and 2012.
Analyzing the temperature values over the years (1960–1990) at Hushyori station, we can say that relatively, several years with relatively hot weather were observed during the considered period: 1963, 1966, 1970, 1971, 1973, 1977, 1979, and 1988. Additionally, the opposite of them was observed in rather cold winter periods of weather: 1964, 1969, 1972, 1974, 1982, 1984, and 1989. For the second period (1991–2018), there were few years with relatively hot weather: 2001, 2007, and 2016. Additionally, the opposite was observed in rather cold winter weather periods: 1991–1993, 1996, and 2012.
Table 2 shows the trend analysis of the average temperature in the upstream, midstream, and downstream areas of the VRB. In the Anzob region (upstream area), according to the observational data in this region, the mean temperature trend in the first period (1961–1990) was recorded in March, September, and November. However, with a relatively large difference in the second period (1991–2018), a considerable increase can be observed in most months of the year. In general, from 1961 to 2018, the average air temperature in January, May, September, October, and December recorded a significant trend. The mean temperature in the Maykhura region in the period from 1964 to 1990 in January and September revealed a significant trend. However, in the next 30 years, that is, from 1991 to 2017, the average air temperature in almost all months, except for August, November, December, and January, has shown a significant trend. In general, during the period from 1964 to 2017, in January, February, March, April, May, and September, an increasing trend in mean temperature was observed. The air temperature in the period from 1960 to 1990 at Hushyori station (downstream region) in January, April, May, September, November, and December shows an upward trend. In February, March, June, July, August, and October, a downward trend was registered (
Table 2). In total, six different months of the year have an upward trend at Hushyori station, and the other 6 months of the year, on the contrary, have a downward trend. During the period 1991–2018, in January, March, April, May, June, and July, a significant increase in air temperature was shown. During the same period, only in November and December, the air temperature recorded a downward trend. From 1960 to 2018, the temperature in March, April, May, and September significantly increased.
Table 3 shows the trend and magnitudinal analysis of the mean annual air temperature at Anzob, Maykhura, and Hushyori stations, which was carried out by Sen’s slope, Pettit’s test, as well as the original and modified MK tests for the trends (Z) in the upstream, midstream, and downstream areas of the Varzob River Basin in Tajikistan during the three time periods. In the Anzob region, during the period from 1961 to 1990, the mean annual temperature indicated a significant upward trend, while during the periods 1991–2018 and 1961–2018, a non-significant positive trend was observed. The analysis of the mean annual temperatures in 1964–1990, 1991–2017, and 1964–2017 exhibited a significant increasing trend in the Maykhura region. In the years from 1991 to 2018, a significant increasing trend in the mean annual temperature was observed at Hushyori station. Similarly, in this region, during the period from 1960 to 2018, we observed an increasing trend for the mean annual temperature.
Figure 4 depicts the change point in the mean annual temperature over three historical periods in the headwaters, midstream, and downstream regions of the Varzob River Basin. The dramatic changes happened in 1976 (in Anzob station), 1976 (in Maykhura station), and 1971 (in Hushyori station) during the first period (1960–1990). For the Anzob, Maykhura, and Hushyori stations for the second period (1991–2018), the rapid changes in mean annual air temperature took place in 2007, 2000, and 1998. The dramatic change occurred for the Maykhura and Hushyori stations in 1998 for the full time (1960–2018) and for the Anzob station in 2007.
4.2. Precipitation Analysis in the Varzob River Basin
Figure 5 indicates the monthly precipitation changes in upstream, midstream, and downstream areas of the VRB. The annual amount of precipitation is 494 mm at Anzob station. The Anzob region is characterized by an annual course of precipitation reaching its minimum in August–September and its maximum in March–May. Annually precipitation mostly (44%) falls in the spring, 26% falls in winter, and 13–17% falls in the summer and autumn months. The average annual relative humidity is 65% and the average monthly ranges from 50 to 80%. Considering the second period (1991–2018), we can say that the period of precipitation has partially changed in temporal precipitation, as there was an increase in precipitation in February and March and a decrease in April. In mountainous terrain, precipitation forms mudflows on the river during the periods of late autumn and during spring floods.
At the Maykhura station, the annual amount of precipitation is about 1185 mm. The Maykhura meteorological station area is characterized by an annual precipitation pattern with a maximum in March (1964–1990) and in February and March (1991–2018) and a minimum in August–September. Most of them fall from December to April. According to the analysis and monitoring of precipitation during the period 1960–1990 at Maykhura station, it can be said that March is considered the most intense month in terms of precipitation, and August is considered the least water. Considering the second period (1991–2018), we can say that the period of precipitation has partially changed in temporal precipitation, as there was an increase in precipitation in February and March and a decrease in April.
At Hushyori station, the annual amount of precipitation is 1188 mm. The station is characterized by an annual course of precipitation with a minimum in August–September and a maximum in March. Their main amount of 45% falls in the spring, 35% falls in the winter, and 5–15% of the annual precipitation falls in the summer and autumn months. The average annual relative humidity is 54% and the average monthly readings range from 38% to 67%. We can note that March is considered the most intense month in terms of precipitation, and August is considered the driest. Considering the second period (1991–2018), it can be said that the period of precipitation has partially changed in the time of precipitation, so in February, there was an increase in precipitation and a decrease in April. The driest month was August.
Considering the period 1960–1990 at Anzob station, one can note that the years with the maximum amount of precipitation were 1963 and 1969, and the year with the least amount of precipitation was 1971. Relative to the obtained graph, it is possible to single out the cyclicity of the repetition of years with unstable precipitation. During the second period 1991–2018, the years with the maximum amount of precipitation were: 1991, 1993, 2004, 2006, 2009, and 2011, while the minimum amount was recorded in the following years: 1996, 2000, 2001, 2002, 2003, 2008, 2010, and 2012–2016 (
Figure 6).
Analyzing the period of 1964–1990 at Maykhura station in the midstream region of the Varzob River Basin, one can note the years with the maximum amount of precipitation—1964 and 1969—as well as the years with the minimum amount of precipitation—1971. During the second period 1991–2018, the years with the maximum amount of precipitation can be noted—1991, 1993, 1998, 2002–2004, 2009, and 2012—as can those with a minimum amount of precipitation—1995, 1996, 2000, 2001, 2008, 2010, and 2015 (
Figure 6). With regard to the obtained graph, one can single out the cyclicity of the repetition of years with unstable precipitation, i.e., sharp fluctuations in precipitation, with strong bursts of years with high precipitation and as opposed to those with the least precipitation. The repetition of years with unstable precipitation, i.e., below normal, ultimately affects the overall flow of the river.
During the period 1960–1990 at Hushyori station, the years with the maximum amount of precipitation were noted to be 1964 and 1969, whilst the year with the minimum amount of precipitation was 1971. During the second period 1990–2018, the years with the maximum amount of precipitation were noted to be 1991, 1993, 1998, 2002–2004, 2009, 2006, and 2013, and the years with the minimum amount of precipitation were 1992, 1995, 1996, 2000, 2001, 2008, 2012, and 2017. Concerning the obtained graph, one can single out the cyclicity of the repetition of years with unstable precipitation, i.e., sharp fluctuations in precipitation, when there were strong bursts of years with high precipitation as opposed to those with the least precipitation.
Table 4 shows the seasonal trend in precipitation in the upstream, midstream, and downstream areas of the VRB. During the period from 1960 to 1990, a significantly increasing precipitation was recorded in the Anzob region only in December and January. During the period from 1991 to 2018, a decreasing trend was recorded in June, July, and December. In general, during the period from 1960 to 2018, a significantly increasing trend was indicated in January and February, and a significantly downward in precipitation was shown in April and May. In the Maykhura station, during the period from 1964 to 1990, precipitation showed a decreasing trend only in April. During the period from 1991 to 2018, a significant trend of decreasing precipitation in June and July was registered, while from 1964 to 2018, a significant trend in decreasing precipitation in April was registered. During the period from 1960 to 1990, at Hushyori station, a significantly downward trend presented in November. From 1960 to 2018, a significant increase in precipitation was recorded in February and June, while a significant decrease in precipitation was only shown in December from 1991 to 2018.
Table 5 presents the results of the annual precipitation trend, slope, and change points in the three meteorological stations in the VRB. In the Anzob station, a significant negative trend in the annual precipitation was found in the upstream region during the periods of 1991–2018 and 1960–2018, while a non-significant positive trend was indicated during the period 1960–1990. According to the conducted analysis, in the Maykhura station (midstream region), precipitation during three historical times periods indicated a significant decreasing trend. Our result for the annual precipitation analysis presented a significant decreasing trend during the period of 1991–2018 at the Hushyori station, while from 1960 to 2018, the annual precipitation was found to have a slightly increasing trend in this region.
Figure 7 depicts the rapid shifts in yearly precipitation in the Varzob River Basin during the three historical periods. At the Maykhura and Hishyori stations, the yearly precipitation experienced dramatic shifts in the first historical time period in 1969, but not at the Anzob station between 1960 and 1990. The dramatic change occurred in the second historical period in 1999 (in Anzob station), 1994 (in Maykhura station), and 2004 (in Hushyori station). Dramatic changes at the Anzob, Maykhura, and Hushyori stations occurred in 1994, 1993, and 1986 during the entire period (1960–2018).
4.3. Snow Cover Analysis in the Varzob River Basin
Here, we described how dependent the river runoff is on snow reserves and trace its maximum runoff and low water period. Let us conduct a small study according to the Anzob meteorological station, which is the only station located in the Varzob river basin at an elevation of more than 3000 m.a.s.l. As can be seen in
Figure 6, the period of maximum snow cover growth occurs in April, when the maximum snow height reaches 178 cm. Note that these data are considered to be the average of three rails installed at the meteorological site. This value is considered optimal for long-term data. There are periods with maximum snow height above two meters (240 cm along one of the rails installed at the meteorological site) and here you need to make allowances for the wind, which plays a role in the horizontal transfer of snow cover.
Before beginning the analysis for this part, it should be noted that we chose the data for 2014 for demonstration purposes due to the availability of snow cover data from the measurement-based weather station for a limited period of time. As shown in
Figure 8a, snow cover is observed all year around for almost nine months and only three months—when the low-water period is considered to be almost zero (with possible precipitation in the form of rain). Thus, we observed that the snow cover appears in early October and practically does not melt after that. Wet-dry snow begins to accumulate in the high-mountainous part of the Varzob River and the fall of this snow does not affect the overall flow of the river. The sub-zero temperature firmly holds the snow crust until the spring, when the warm rays of light begin to saturate the snow cover with warmth in the daytime and “squeeze out” the resulting water at night. The flow of the river during this period is considered the most complete and full-flowing. The coldest months were February and December at Anzob station with a mean daily temperature of −15 °C, and the warmest temperature was recorded in July at +10.1 °C. At the same time, the snow cover lay until June, i.e., it can be said that the last natural recharge of the river runoff ends in June and precipitation in the form of rain wash away the remnants of the snow cover. The accumulation of snow cover occurs from October and continues until the beginning of April, and then, precipitation in the form of rain begins to wash away the accumulated snow cover, which has a beneficial effect on the runoff of the Varzob River. During the low-water period (July, August, September), the precipitation does not have a special effect on the increase in the river flow.
The Maykhura snow avalanche station in the midstream region (
Figure 8b) is the only station located in the Varzob River Basin at an elevation of 2000 m.a.s.l. As shown in
Figure 6, the maximum height in snow cover occurs in March, when the maximum snow height reaches almost 138 cm. Snow cover is observed all year for almost six months and only six months—the low-water period is considered to be almost zero (with the possible loss of small precipitation in the form of rain). Thus, we note that the snow cover appears in the first ten days of November and is practically unstable. Wet-dry snow begins to accumulate in December when the temperature is firmly in minus weather. The sub-zero temperature firmly holds the snow crust until the onset of spring. The coldest months at Maykhura station were January, with a mean daily temperature of −6.1 °C, and the warmest temperature was recorded in July at +19.4 °C. At the same time, the snow cover lay until May, i.e., it can be said that the last natural recharge of the Varzob river runoff ends in May.
We analyze the period of the accumulation of snow cover in the lower parts of the Varzob river. At Hushyori station, the maximum increase in snow cover occurs in February, when the maximum snow height reaches almost 54 cm (
Figure 8c). These data are considered the average value for three rails installed at the meteorological site. Snow cover is observed all year around for almost five months and the low-water period is considered to be almost zero (with possible slight precipitation in the form of rain) for only seven months. Snow cover appears in the last ten days of November and is practically unstable. Wet-dry snow begins to accumulate in December when the night temperature is firmly in below zero.
At Hushyori station, the temperature is below zero only in February—this is the coldest month with a mean daily temperature of −2.4 °C, and the warmest temperature was recorded in July at +24.2 °C. At the same time, the snow cover lay until April, i.e., it can be said that the last natural recharge of the river flow ends in March in VRB and the observed precipitation in the form of rain washes away the remnants of the snow cover. The highest snow is observed in the high-mountain zone (Anzob area) of the Varzob River Basin, with a maximum snow height exceeding 2.5 m. Snow saturation occurs in winter during unstable weather and fog formation, as well as in February–March at positive daytime temperatures.
Ackroyd et al. [
61] studied the snow cover dynamics in the Amu Draya and Syr Darya River Basin in Central Asia. Their study indicated that the trends along the Syr Darya River and Amu Darya River did not fully apprehend the inconsistency across various features and other topographical factors. An increasing rate of warming does have a greater effect at higher elevations, according to the trends analysis across elevation bands [
62]. Nonetheless, there is still potential to determine more localized trends within each basin to better understand the spatial variability in mountainous regions. This study applied the snow cover data from the MODIS in the Varzob River Basin in Amu Darya River Basin. It should be noted that MODIS images have been extracted which have less than a 10 percent cloud. The MODIS satellite data are available for the period of 2001–2022, which is used to analyze snow cover in the region of interest. Therefore, the study is limited to this time period as it is the only period for which the necessary data are available.
Figure 9 shows the MK trend test based on the p-value. The
p-value is a statistical metric that helps define whether the observed snow cover trend is more likely to be a real and significant trend or just the result of chance or random variation in the data. The p-value for our study was found to be less than 0.05, which means that there is a 95% probability that the observed trend in snow cover is real and significant rather than the result of chance or random variation in the data. Accordingly, the snow cover trend is regarded as statistically significant at a 95% confidence level [
63].
The MK test was applied to analyze the snow cover area trend.
Figure 9 shows the overall monthly snow cover trends of the study area. Our results indicate the overall decreasing trend in snow cover area during February, September, and November and an increasing trend in July. We further analyzed the snow cover trends at three available station-based data and describe these results in
Table 6. We observed that, at a height elevation area, the snow cover area showed a decreasing trend spatially in the winter season. Hushyori station belongs to the low elevated region and showed the decreasing trend during most of the months. These results explain the clear picture of the snow cover area overall, and at the evolution level, show trends on a monthly basis in the VRB. The decreasing trend in February and September could be more of a concern about the hydrological system of the study area because these months relate to the season shifting month, and the trend results indicate the intensity of the warming conditions in VRB during the period 2001–2022.
In our study area, the heatmap reveals some interesting trends. Understanding the snow cover dynamics can be helped by looking at the heatmap, which provides a visual depiction of the patterns and changes in snow cover area over time and across several months (
Figure 10). The heatmap demonstrates that, as is typical for most places, the maximum amount of snow cover is in winter (December–February) and the minimum is in summer (June–August). Likewise, the amount of snow cover over the period represented in our heatmap may vary slightly from year to year. The X axis also appears to indicate some variation in snow cover from year to year, with some years having higher snow cover than others. The spatial patterns of snow cover also vary across the years. For instance, some years show a more widespread snow cover over the study region, while the other years show a sparser snow cover. In the Varzob River Basin, the sum of the snow cover area from January to December was found for 13,013 km
2 in 2014 and 9063 km
2 in 2007, which are considered the maximum and minimum snow cover zones during the period 2001–2022, respectively. In general, seasonal snow cover changes during the study period (2001–2022) indicated the largest area covered by snow in January, February, March, and December, and showed the smallest area of snow cover in June, July, and August.
Figure 10 indicates that there appears to be a pattern in snow cover throughout the year, as well as across the 22 years from 2001 to 2022. There is maybe a connection between these patterns and variables including the temperature and precipitation patterns.