The alternation of seasons is caused by the Earth’s revolution and solar radiation, which lead to the latitudinal and altitudinal heat redistribution and cold-hot alternation (
Journal of Geographical Sciences, Volume. 30, Issue 9, 1387(2020)
Spatio-temporal variation in China’s climatic seasons from 1951 to 2017
In this paper, meteorological industry standard, daily mean temperature, and an improved multiple regression model are used to calculate China’s climatic seasons, not only to help understand their spatio-temporal distribution, but also to provide a reference for China’s climatic regionalization and crop production. It is found that the improved multiple regression model can accurately show the spatial distribution of climatic seasons. The main results are as follows. There are four climatic seasonal regions in China, namely, the perennial-winter, no-winter, no-summer and discernible regions, and their ranges basically remained stable from 1951 to 2017. The cumulative anomaly curve of the four climatic seasonal regions clarifies that the trend of China’s climatic seasonal regions turned in 1994, after which the area of the perennial-winter and no-summer regions narrowed and the no-winter and discernible regions expanded. The number of sites with significantly reduced winter duration is the largest, followed by the number of sites with increased summer duration, and the number of sites with large changes in spring and autumn is the least. Spring advances and autumn is postponed due to the shortened winter and lengthened summer durations. Sites with significant change in seasonal duration are mainly distributed in Northwest China, the Sichuan Basin, the Huanghe-Huaihe-Haihe (Huang-Huai-Hai) Plain, the Northeast China Plain, and the Southeast Coast.
1 Introduction
The alternation of seasons is caused by the Earth’s revolution and solar radiation, which lead to the latitudinal and altitudinal heat redistribution and cold-hot alternation (
China is vulnerable to climate change due to the vast land and uneven terrain. Some recent studies have revealed that the average temperature in China is rising faster than the global average (
In this study, an improved multiple regression method is used to calculate the spatial distribution of climatic seasonal regionalization. The principal types of climatic seasons present in China are also revealed. At the same time, the climatic season classification criteria are utilized to calculate the trends of the duration and start dates of the climatic seasons. In this paper, the spatio-temporal distribution of China’s climatic seasons is analysed in detail. Additionally, a reference for the meteorological service industry is presented.
2 Data
The datasets of daily mean surface air temperature in China are collected separately from the China Meteorological Administration (
Figure 1.
3 Methods
3.1 Regionalization of the climatic seasons
Seasonal temperature thresholds are defined variously in different regions. In this study, the approach proposed by climatic seasonal division standard (QX/T 152-2012) is used to classify China’s climatic seasons, and it has been widely promoted by the Chinese meteorological department. A few concepts are included as follows:
3.1.1 Normal climatic seasons
The normal climatic seasons refer to seasons during Climate Normals. The start dates, end dates and duration of normal climatic seasons are determined by the five-point smoother temperature of the perennial series. The formulas are as follows:
where $T{{\overline{M}}_{j}}$(℃) is the five-point smoother temperature of the perennial series on the
Threshold temperatures for different climatic seasons
Threshold temperatures for different climatic seasons
|
Normal climatic seasonal areas can be defined by the durations of the climatic seasons. When the climatic seasonal duration is more than 5 days in an area, the season is stable.
Definition of normal climatic seasonal regions
Definition of normal climatic seasonal regions
|
3.1.2 Annual climatic seasons
If the start date of a season calculated for the first time is over 15 days earlier than the normal climatic seasonal date, a second judgement should be made. If the initial 5-day continuous sliding average temperature sequence does not meet the season’s indicator, the sequence must be calculated again for them to meet. When the cumulative number of days that meet the season’s indicator is greater than or equal to the number of unsatisfied days during two consecutive processes, the start date of the initial judgement is taken as the start date of the climatic season. Otherwise, the second judgement will be taken.
3.2 Simulation of climatic seasonal spatial distribution
The spatio-temporal continuity of temperature is the best among all the meteorological elements. There are linear relationships between temperature spatial distribution and altitude, latitude and longitude (
where
The distribution of climatic seasons is determined by the climatic seasonal duration. The spatial distribution formula of the climatic seasonal duration is obtained by a multiple regression model. The formula in
Multiple regression models for climatic seasonal length simulation in China for different Climate Normals
Multiple regression models for climatic seasonal length simulation in China for different Climate Normals
|
The accuracy of the multiple regression simulation is verified by using the correlation coefficient (
The R, S and RMSE between measured and simulated values of the validation site
The R, S and RMSE between measured and simulated values of the validation site
|
4 Results
4.1 Spatial distribution of the normal climatic seasons
The spatial distribution of climatic seasons in the four Climate Normals (1951-1980, 1961-1990, 1971-2000, and 1981-2010) is calculated by multiple regression models.
The seasonal variation in area for the climatic seasons among different Climate Normals is not obvious. It can be seen from
Figure 2.
4.2 Spatial distribution of climatic seasonal duration in different Climate Normals
The durations of spring, summer, autumn and winter in four Climate Normals are counted in
Figure 3.
4.3 Variations of the climatic seasonal distribution from 1951 to 2017
The method in 3.1.2 is used to determine the climatic seasonal regions for each year from 1951 to 2017, and the proportion of each subregion to China is calculated. The results indicate that the PWR proportion is decreasing, and the NWR, NSR and DR are increasing with the linear trends and fitting formulas in
Figure 4.
The cumulative anomaly curves of the climatic seasonal division areas, based on the 1981-2010 Climate Normals, are calculated to evaluate the area variational trend. A similar situation can be found in four cumulative anomaly curves in 1994. The PWR and NSR areas expanded before 1994, while the NWR and DR narrowed. However, the opposite situation occurred after the turning point in 1994. The following characteristics can be found in
Figure 5.
4.4 Variations of climatic seasonal start dates and durations from 1951 to 2017
Through previous analyses, the trend of each seasonal duration from 1951 to 2017 is calculated in
Figure 6.
The methods in 3.1.1 and 3.1.2 are used to calculate the start date of spring, summer, autumn and winter each. Sen’s slope estimator (
There are 327 sites with statistically significant differences in the spring start date (
Figure 7.
5 Conclusions
In this paper, a criterion for climatic seasonal divisions is used to analyse the spatio-temporal variations of China’s climatic seasons from 1951 to 2017. To use the data from more sites in calculating the climatic seasons, the multiple linear regression method is applied to compensate for the non-recorded average daily temperature. The multiple regression model used to denote the spatial distribution of meteorological elements is also improved. Some highlights in the above analyses are as follows:
(1) After the selected site is verified, the temperature data of the vacancy and the improved multiple regression method can be used to represent the spatial and temporal distribution of climatic seasons in China.
(2) There are four division types for China’s climatic seasons, namely, the PWR, NWR, NSR and DR. The area of NWR, PWR, NSR and DR respectively accounts for 4%, 10.5%, 27.5% and 58% of the China’s land area.
(3) According to the analyses of four Climate Normals and the multi-year climatic seasonal regions, it can be seen that there has been no significant variation in China’s climatic seasonal regions from 1951 to 2017.
(4) The cumulative anomaly curve of the four climatic seasonal areas suggests that the trend of China’s climatic seasonal regions turned in 1994, after which the area of the PWR and NSR narrowed by 29% and 11%, and NWR and DR expanded.
(5) The number of sites with significantly shortened winter duration is the largest, followed by sites with lengthened summer duration, and the number of sites with significant changes in spring and autumn is the least.
(6) The spring comes sooner and the autumn is postponed due to shortened winter and lengthened summer duration.
(7) Sites with significant changes in season duration are mainly distributed in Northwest China, the Sichuan Basin, the Huang-Huai-Hai Plain, the Northeast China Plain, and the Southeast Coast.
Compared with previous studies of multi-climate factors, it is reasonable for this paper to choose only the daily average temperature to explain the large-scale climatic seasonal distribution and climatic regions. In China, with diverse climate types, a unified definition is needed for weather forecasting and monitoring, as well as serving production and people’s daily lives.
[8] et alIntroduction. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Computational Geometry, 18, 95-123(2013).
[9] et alChanges of annual accumulated temperature over southern China during 1960-2011. Journal of Geographical Sciences, 25, 1155-1172(2015).
[10] Response of the starting dates and the lengths of seasons in Mainland China to global warming. Climatic Change, 99, 81-91(2010).
[11] Dates of frost onset, frost end and the frost-free season in Turkey: Trends, variability and links to the North Atlantic and Arctic Oscillation indices, 1950-2013. Climatic Change, 69, 155-176(2016).
[13] On the estimation of annual and monthly mean values of air temperature over Greece using stepwise multiple regression analysis. Archives for Meteorology Geophysics and Bioclimatology Series B-Theoretical and Applied Climatology, 32, 287-295(1983).
[14] Downscaled estimates of late 21st century severe weather from CCSM3. Climatic Change, 129, 307-321(2015).
[15] Space-time variations of climatic seasons and their correlation with the phenological development of nature in Estonia. Climatic Change, 15, 207-219(2000).
[16] et alSpatial and temporal variability of climatic seasons on the East European Plain in relation to large-scale atmospheric circulation. Climate Research, 23, 111-129(2003).
[18] Coupled spatiotemporal variability of temperature and spring phenology in the Eastern United States. International Journal of Climatology, 36, 1744-1754(2016).
[19] et alSpatiotemporal change in China’s climatic growing season: 1955-2000. Climatic Change, 99, 93-118(2010).
[22] Substantial increase in concurrent droughts and heatwaves in the United States. Proceedings of the National Academy of Sciences of the United States of America, 112, 11484-11489(2015).
[23] Analysis of the division of four seasons in China and its changing characteristics, 693-701(2007).
[24] China’s National Assessment Report on Climate Change(2015).
[25] et alResponse of changes in seasonal soil freeze/thaw state to climate change from 1950 to 2010 across China. Journal of Geophysical Research Earth Surface, 121, 1984-2000(2016).
[26] Estimates of the regression coefficient based on Kendall’s Tau. Publications of the American Statistical Association, 63, 1379-1389(1968).
[29] et alAssessing the effects of climate change on the phenology of European temperate trees. Agricultural and Forest Meteorology, 151, 969-980(2011).
[30] et alSpring temperature change and its implication in the change of vegetation growth in North America from 1982 to 2006. Proceedings of the National Academy of Sciences of the United States of America, 108, 1240-1245(2011).
[31] Accounting for autocorrelation in detecting mean shifts in climate data series using the Penalized Maximal t or F Test. Journal of Applied Meteorology & Climatology, 47, 2423-2444(2007).
[33] Penalized maximal t test for detecting undocumented mean change in climate data series. Journal of Applied Meteorology and Climatology, 46, 916-931(2007).
[35] et alSpatial and temporal variation characteristics of the four seasons in the northern China. Scientia Geographica Sinica, 29, 1105-1110(2011).
Get Citation
Copy Citation Text
Bin MA, Bo ZHANG, Lige JIA. Spatio-temporal variation in China’s climatic seasons from 1951 to 2017[J]. Journal of Geographical Sciences, 2020, 30(9): 1387
Category: Research Articles
Received: Dec. 17, 2019
Accepted: Jun. 2, 2020
Published Online: Apr. 21, 2021
The Author Email: ZHANG Bo (zhangbo@nwnu.edu.cn)