Introduction

The Qinling–Daba mountains (QDM), located in the North–South transition zone and between the Tibetan Plateau and the eastern plains in China, is considered both as a natural boundary between the warm temperate zone and the north subtropical zone, and as an ecological corridor connecting the Tibetan Plateau and the eastern plains1,2. The climate of the QDM is accordingly characterized by complexity and diversity on different flanks and terrain3,4,5. The southern flank of the Qinling Mountains, presenting windward slopes to moisture-carrying air from the southeast, has a sunny aspect, relatively flat terrain affected by fold belts, and wet and warm climatic conditions, whereas the northern flank, presenting leeward slopes, has a shady aspect, steep terrain formed by fault belts, and relatively dry and cold climatic conditions1,6. Accordingly, climate change and its influence vary markedly on different flanks and at different elevations. It has been recognized that the humidification rate and climate warming are greater on the northern flank of the Qinling Mountains than on the southern flank7,8, but that the improvement in net primary production of vegetation is higher on the southern flank than on the northern flank at elevations below 2000 m4,7. However, most previous related studies focused mainly on climate change in partial regions, e.g., the Qinling Mountains3,7,9, Huaihe Basin10, and Daba Mountains11. Although the QDM represent an integrated geographic region connecting the warm temperate zone to the north subtropical zone, little attention has been paid to climate change within the entire region1. The objective of this study was to reveal the spatiotemporal variation of climate change and its complexity on different flanks and at different elevations in the QDM, to contribute to the exploration of the cause of the North–South regional differentiation in China. First, we collected and compiled air temperature data and precipitation data measured at 118 national weather stations (1969–2018). Then, we used the Manne–Kendall method to analyze the heterogeneities and similarities with climate change on different flanks and at different elevations in the QDM during 1969–2018.

Study area

The QDM, located in central China (30°–36°N, 101°–115°E, span six provinces: Gansu, Sichuan, Shaanxi, Chongqing, Henan, and Hubei (Fig. 1). The QDM, consisting primarily of the Qinling Mountains and the Daba Mountains separated by the Han River valley, form a natural boundary between the warm temperate zone and the subtropical zone. Generally, the Qinling Mountains are broadly coincident with the 0 °C isotherm in January, the 800-mm isohyet, and the 2000-h sunshine duration contour in China12. Variation in these climatic factors causes the regional vegetation to change gradually from a deciduous broadleaved forest zone to an evergreen broadleaved forest zone across the QDM from north to south13,14. However, western parts of the QDM are dominated by cold temperate grassland because of the higher elevation and the drier climate in certain valleys15.

Figure 1
figure 1

The study area and the spatial distribution of weather stations in the QDM. Pink dots represent weather stations on the southern flank of the Qinling Mountains (31), orange dots represent weather stations on the northern flank of the Qinling Mountains (20), purple dots represent weather stations on the southern flank of the Daba Mountains (13), red dots represent weather stations on the northern flank of the Daba Mountains (20), and blue dots represent weather stations on the western flank of the Qinling–Daba mountains (34). Notes: the map is created via the software ArcGIS 10.2 and the related research marks and words are added. The data of the map is from the site: https://earthexplorer.usgs.gov/, https://data.cma.cn/.

Full size image

A network of 118 national weather stations is distributed over different flanks in the QDM: 31 and 20 stations are located on the southern and northern flanks of the Qinling Mountains, respectively, 13 and 20 stations are distributed on the southern and northern flanks of the Daba Mountains, respectively, and 34 stations are distributed on the western flank of the QDM (Fig. 1).

Materials and methods

Weather stations

A dataset comprising daily temperature and precipitation data (1969–2018) measured at 118 national weather stations in the QDM was obtained from the National Meteorological Information Center (https://data.cma.cn/), and the temperature and precipitation data were averaged by year, season and month. As shown in Fig. 1, the weather stations are mainly distributed over six provinces: Gansu, Shaanxi, Henan, Sichuan, Hubei and Chongqing.

DEM data

For this study, digital elevation model (DEM) data for different blocks of the QDM were downloaded from the website of the U.S. Geological Survey (https://earthexplorer.usgs.gov/) and then spliced into a complete regional DEM covering the entire area of the QDM with 1000-m spatial resolution. To analyze the association between climate change and elevation, the DEM data were classified into four levels: 0–1000, 1000–2000, 2000–3000, and 3000–4000 m. The variation characteristics of the weather stations (1969–2018) were determined for the different classification segments, and the relationship between climate change and elevation specific to the QDM was explored.

Exposure data

To decipher the association between climate change and different flanks in the QDM, we divided the QDM into five parts: the southern and northern flanks of the Qinling Mountains, the southern and northern flanks of the Daba Mountains, and the western flank of the QDM. We classified and divided the QDM on the basis of the positions of the main ridge lines and major rivers (i.e., the Jialing River and the Han River) shown in Fig. 1. The rates of change of temperature and precipitation in the five regions were calculated through statistical analysis to explore the variations and similarities of climate change on the different flanks of the QDM in the North–South transition zone of China.

Method

The widely used Manne–Kendall method is effective in detecting change in long-term time series16. The details of the method of treatment are described in Xu et al.17 In this study, the Manne–Kendall method was applied to detect changes in the long-term trends of temperature and precipitation on different flanks and at different elevations of the QDM using the meteorological data from the 118 national weather stations.

Results

Air temperature changes on different flanks in the QDM during 1969–2018

The Mann–Kendall test revealed that 108 (91.53%) of the stations presented a significant trend of increase in spring air temperature (Fig. 2), while 61 (51.69%), 91 (77.12%), and 102 (86.44%) of the stations showed a substantial trend of increase in air temperature in summer, autumn, and winter, respectively. A significant trend of increase in air temperature was shown by a greater proportion of stations on the western flank of the QDM than by stations on other flanks in different seasons. For example, 97.06%, 88.24%, 94.12%, and 97.06% of stations in the western QDM were found to have a notable trend of increase in air temperature in spring, summer, autumn, and winter, respectively, whereas such trends were detected on the other flanks at less than 89.29%, 35.71%, 69.05%, 80.95% of stations, respectively. In comparison with the southern flank of the Qinling Mountains, the northern flank had a greater proportion of weather stations showing a significant trend of increase in air temperature in different seasons. For example, 95%, 40%, 75%, and 95% of stations on the northern flank showed a significant trend of increase in spring, summer, autumn, and winter, respectively, while in the corresponding seasons, 93.55%, 35.48%, 64.52% and 87.10% of stations on the other flanks showed a significant trend of increase, respectively. On the southern flank of the Daba Mountains, 7.69% of stations showed a significant negative trend in air temperature in summer, autumn, and winter.

Figure 2
figure 2

Significance analysis of mean air temperature based on the Mann–Kendall test at the 5% significance level at different weather stations in the QDM during 1969–2018. Notes: the map is created via the software ArcGIS 10.2 and the related research marks and words are added. The data of the map is from the site: https://earthexplorer.usgs.gov/, https://data.cma.cn/.

Full size image

Trend analysis for the mean air temperature in different seasons revealed a significant trend of increase on the different flanks of the QDM during 1969–2018 (Fig. 3, Table 1). However, the rate of increase in air temperature in different seasons varied markedly between different flanks. The rate of increase in air temperature on the northern flank of the QDM was greater than that on the southern flank in all four seasons (Fig. 3). Specifically, the air temperature tendency on the different flanks of the QDM was greater in spring and winter than in summer and autumn. Climate warming was least in summer, i.e., the rate of increase in air temperature was as low as 0.123, 0.074, 0.085, 0.056, and 0.285 °C/10a on the northern flank of the Qinling Mountains, the southern flank of the Qinling Mountains, the northern flank of the Daba Mountains, the southern flank of the Daba Mountains, and the western flank of the QDM, respectively.

Figure 3
figure 3

(From top row to bottom row) Variation in mean air temperature on the southern flank of the Qinling Mountains, the northern flank of the Qinling Mountains, the southern flank of the Daba Mountains, the northern flank of the Daba Mountains, and the western flank of the Qinling–Daba mountains (a1–a5 spring), (b1–b5 summer), (c1–c5 autumn), and (d1–d5 winter).

Full size image
Table 1 Seasonal air temperature trends (°C/10a) on different flanks in the Qinling–Daba mountains during 1969–2018.
Full size table

Air temperature changes at different elevations in the QDM during 1969–2018

Statistical analysis of the trends in mean air temperature showed that stations with significant trends of increase varied markedly in different seasons and at different elevations in the QDM (Fig. 2). In the QDM, 79 (89.77%), 31 (35.23%), 61 (69.32%), and 72 (81.82%) of the stations were found to have a significant trend of increase in air temperature at elevations of 0–1000 m in spring, summer, autumn, and winter, respectively, whereas 98.68%, 92.50%, and 100% of stations were found to have a significant trend of increase at elevations of 1000–2000, 2000–3000, and 3000–4000 m, respectively. It was discovered that the area of warming in high-elevation areas was greater than that in low-elevation areas in different seasons.

The air temperature trend analysis for different seasons revealed significant trends of increase at different elevations in the QDM during 1969–2018. (Fig. 4, Table 2). However, the rate of increase in mean air temperature varied greatly depending on season and elevation. At higher elevations, we detected higher rates of increase in mean air temperature in summer, autumn, and winter (Fig. 4, Table 2). For example, at elevations of 3000–4000 m, the rate of increase in air temperature in summer, autumn, and winter could reach 0.440, 0.390, and 0.456 °C/10a, respectively, while at elevations of 0–1000 m, the rate of increase was 0.205, 0.218, and 0.303 °C/10a, respectively. Conversely, the rate of increase in air temperature in spring was higher in low-elevation areas (0–1000 and 1000–2000 m) than in high-elevation areas (2000–3000 and 3000–4000 m).

Figure 4
figure 4

Seasonal rate of increase in air temperature at different elevations in the Qinling–Daba mountains during 1969–2018.

Full size image
Table 2 Seasonal rate of increase in mean air temperature (°C/10a) at different elevations in the Qinling–Daba mountains during 1969–2018.
Full size table

Precipitation changes on different flanks and at different elevations in the QDM during 1969–2018

The Mann–Kendall tests did not indicate significant trends in precipitation (Fig. 5). However, some stations did exhibit notable trends in precipitation in spring, summer, and winter (Fig. 6). The weather stations are distributed at various elevations on the different flanks of the QDM. For example, Mount Hua station is in the 2000–3000 m zone on the northern flank of the Qinling Mountains, Shiquan station is in the 0–1000-m zone on the southern flank of the Qinling Mountains. Taibai station is in the 1000–2000-m zone on the southern flank of the Qinling Mountains, Zigui station is in the 0–1000-m zone on the southern flank of the Daba Mountains, and Beichuan station is in the 0–1000-m zone and Diebu, Songpan, and Hezuo stations are in the 2000–3000-m zone on the western flank of the QDM.

Figure 5
figure 5

(From top to bottom) Variation in mean precipitation on the southern flank of the Qinling Mountains, the northern flank of the Qinling Mountains, the southern flank of the Daba Mountains, the northern flank of the Daba Mountains, and the western flank of the Qinling–Daba mountains (a1–a5 spring, b1–b5 summer, c1–c5 autumn, and d1–d5 winter).

Full size image
Figure 6
figure 6

Significance analysis of precipitation changes for different weather stations based on the Mann–Kendall tests at the 5% significance level at different elevations, on different flanks, and in different seasons in the Qinling–Daba mountains during 1969–2018.

Full size image

As shown in Fig. 7 and listed in Table 3, Zigui and Shiquan stations showed significant trends of increase in summer precipitation, with change rates as high as 47.36, and 22.91 mm/10a, respectively, while one station (Diebu) showed a significant trend of decrease (− 19.43 mm/10a). In contrast, four weather stations (Zigui, Beichuan, Taibai, and Hezuo) indicated increasing changes in winter precipitation. It is important to note that Mount Hua station exhibited a significant trend of decrease in precipitation in both spring and autumn with a rate of − 14.60 and − 11.50 mm/10a, respectively. In spring, only Songpan station showed a trend of increase in precipitation, i.e., 7.63 mm/10a.

Figure 7
figure 7

Stations with significant change in precipitation in different seasons in the Qinling–Daba mountains during 1969–2018: (a) spring, (b) summer, (c) autumn, and (d) winter.

Full size image
Table 3 Seasonal rate of change in mean precipitation (mm/10a) at weather stations with a significant trend in different seasons in the Qinling–Daba mountains during 1969–2018.
Full size table

Discussion

Our analysis confirmed that air temperature in the QDM has exhibited a trend of increase over the previous 50 years, but that the rate of temperature increase presents marked variation between different flanks and at different elevations. Overall, the rate of change of air temperature was greater on the northern flank and at higher elevations than on the southern flank and at lower elevations of the QDM in all seasons. For example, the rate of change of air temperature over the previous 50 years was as high as 0.374 and 0.456 °C/10a on the northern flank of the Qinling Mountains and at 3000–4000 m in the QDM, respectively, but as low as 0.267 and 0.303 °C/10a on the southern flank of the Qinling Mountains and at 0–1000 m in the QDM, respectively. A similar result was also identified in relation to the Qinghai–Tibet Plateau by Liu and Yao18,19. Additionally, research in the Tianshan Mountains, Swiss Alps, and Colorado Rockies by Xu et al.20, Beniston et al.21 and Rangwala et al.22, respectively, also found that the rate of climate warming in high-elevation areas has been higher than that in low-elevation areas. A possible reason is that air temperature change on northern flanks and at relatively high elevations is more sensitive to driving factors such as snow/ice coverage22, cloud18,23, water vapor24,25, carbon dioxide26,27, and soil moisture28.

The findings of this study highlighted the variation in air temperature increase in different seasons. In comparison with low-elevation areas, the rate of increase in air temperature was higher at high elevations in summer, autumn, and winter. Conversely, in spring, it was higher at lower elevations than at higher elevations. Dong Danhong et al.29 also found that the change in the rate of increase in air temperature at lower elevations was greater than that at higher elevations in spring. Possible causes of this phenomenon include human activities30 and solar radiation31,32. However, clarification of how these factors might affect the change in air temperature at different elevations in spring needs further research. The air temperature tendency on different flanks of the QDM was greater in spring and winter than in summer and autumn. The findings of this study indicated that the increase in air temperature was more pronounced in spring and winter, consistent with the results of Gao and Zhu et al. regarding air temperature change in the QDM and on the Qinghai–Tibet Plateau, respectively8,33. Plausible explanations for these results include the normalized difference vegetation index34, latent heat generated by local precipitation35, and air temperature inversion due to topography36. Anthropogenic factors such as surface albedo variation associated with land use change could explain the variation of air temperature in different seasons in mountain regions2,37, and the combined effects of these factors could account for the diversity of seasonal air temperature in the interior of mountain systems with complex topography.

Precipitation did not change significantly in most of the study area during 1969–2018, which is a finding consistent with the results of Wang et al. regarding the spatiotemporal changes of extreme precipitation in the North–South transition zone of China during 1960–201738. However, eight weather stations did show a consistent tendency of increase in precipitation in winter. This means that the climate has become wetter in winter, which has helped alleviate winter droughts in the QDM, especially in the western QDM. Our results are in agreement with the findings of Gou Xiaoxia30 and Gao Xiang8 in relation to the northern slopes of the Tianshan Mountains and the QDM, respectively. Wetter winters might play an important role in relation to changes of vegetation, phenology39,40, net primary production41, and the normalized difference vegetation index42, which should be studied further in the future.

The seasonal rates of air temperature increase that differ between the various subregions and at different elevations in the QDM, will affect changes in vegetation cover and biodiversity. The rate of air temperature change was found to be greater on the northern flank than on the southern flank, and greater at higher elevations than at lower elevations. Therefore, it will result in faster increase in vegetation coverage at higher elevations on the northern flank than that at lower elevations on the southern flank, as verified by Kullman43 and Walther44. Accordingly, carbon sequestration might be greater at higher elevations than at lower elevations, which has also been verified by recent research45. In the future, high-elevation vegetation might migrate upward, resulting in increased species diversity.

Conclusions

We used climate statistical analysis methods to analyze the spatiotemporal trends of air temperature and precipitation during 1969–2018 on different flanks and elevations in the QDM. The following conclusions were obtained.

  1. (1)

    Overall, an obvious trend of warming in different seasons has occurred over the previous 50 years in the QDM, but the rate of air temperature increase has varied markedly on different flanks and at different elevations. Precipitation has not changed significantly in most areas of the QDM during 1969 – 2018, except at eight stations that showed a significant change in precipitation to a certain degree in different seasons.

  2. (2)

    The seasonal rate of increase in air temperature presented notable variations on the different flanks of the QDM. Generally, the rate of air temperature change was greater on the northern flank than on the southern flank in all seasons in the QDM. Specifically, the air temperature tendency was greater in spring and winter than in summer and autumn on the different flanks in the QDM.

  3. (3)

    There was an increasing trend of seasonal air temperature at different elevations in different seasons in the QDM. With increasing elevation, the rate of increase in mean air temperature showed a tendency of increase in summer, autumn and winter, whereas the rate of increase in mean air temperature in spring was higher at lower elevations than at the higher elevations.