Advertisement

International Journal of Disaster Risk Science

, Volume 7, Issue 3, pp 216–226 | Cite as

World Regionalization of Climate Change (1961–2010)

  • Peijun Shi
  • Shao Sun
  • Daoyi Gong
  • Tao Zhou
Open Access
Article

Abstract

Traditional climate classification or regionalization characterizes the mean state of climate condition, which cannot meet the demand of addressing climate change currently. We have developed a climate change classification method, as well as the fundamental principles, an indicator system, and mapping techniques of climate change regionalization. This study used annual mean temperature and total precipitation as climatic indices, and linear trend and variation change as change indices to characterize climate change quantitatively. The study has proposed a scheme for world climate change regionalization based on a half century of climate data (1961–2010). Level-I regionalization divides the world into 12 tendency zones based on the linear trend of climate, level-II regionalization resulted in 28 fluctuation regions based on the variation change of climate. Climate change regionalization provides a scientific basis for countries and regions to develop plans for adapting to climate change, especially for managing climate-related disaster or environmental risks.

Keywords

Climate change Regionalization Temperature Precipitation Tendency Fluctuation World 

1 Introduction

Climate condition is the foundation for the sustainable development of human society. Since the Industrial Revolution 200 years ago, many of the observed climate changes are unprecedented over millennia, and have begun to influence the natural environment seriously IPCC 2012, 2013). Addressing climate change has become a global issue that has attracted the attention of politicians, scientists, and the public worldwide. We believe that climate change adaptation should be in line with regional and local conditions, that is, based on an analysis of the relationship between different climate change types and related disasters or environmental risks.

Traditional climate classifications and regionalization schemes characterize the mean state of climate condition (Köppen 1936; Thornthwaite 1948; Strahler 1975) for which the indices are the mean value of climate variables over a certain period of time. With climate change accelerating nowadays, these traditional works cannot meet the demand of climate change adaptation. In the past century, global warming was the main feature of climate change; however, the rate of temperature increase varied greatly from high latitudes to equatorial areas, while temperature was even cooling in the Andes Mountains of South America. Other climatic indices including precipitation, potential evapotranspiration, wind speed, and so on also showed regional differentiation patterns on different temporal scales, and the features were more complicated (Dai et al. 1997; Emori and Brown 2005; Hansen et al. 2010; Wu et al. 2011; Ji et al. 2014).

We had developed a set of classification methods that identify regional differences of climate change, and had used these methods to complete the climate change regionalization in China (Shi et al. 2014). This study used the same methods and global gridded datasets of temperature and precipitation from 1961 to 2010 to recognize climate change quantitatively, and puts forward a scheme of climate change regionalization on a world scale with administrative subdivisions of countries and regions as the basic unit.

2 Data and Methods

Data used in the research mainly include meteorological data, topographic data, and administrative boundary data. Change in climatic variables was divided into linear trend and variation change, that is, tendency and fluctuation. This article uses annual mean temperature and annual precipitation as the two most important climatic variables with which to diagnose climate change. The confidence of linear regression is calculated by the F-test method during the calculation of indices to evaluate the reliability of the results.

2.1 Data Sources

Temperature data are derived from the Monthly Dataset (0.5° resolution, from January 1961 to December 2010) of the Climatic Research Unit, University of East Anglia (Mitchell and Jones 2005), precipitation data are from the Monthly Dataset (0.5° resolution, from January 1961 to December 2010) of Global Precipitation Climatology Centre1 (Schneider et al. 2011). Topographic data are from Global 30 Arc-Second Elevation dataset provided by the United States Geological Survey.2 Administrative boundary data are from Global Administration database, developed by Robert Hijmans, University of California, Davis.3

2.2 Calculation Methods

Climate change can be classified into nine modes based on tendency (rise, decline, not significant) and fluctuation (increase, decrease, not significant) of the climatic variables (Shi et al. 2014), as shown in Fig. 1.
Fig. 1

Modes of climate change

For the original series x i , considering the disturbance of interannual scale variation, first apply the moving average method to establish a 5-year moving average series y i ; t i represents the corresponding time. Then develop a linear regression equation between y i and t i :
$$\hat{y}_{i} = a + bt_{i} .$$

The regression coefficient b is called the “tendency value,” which can be calculated through the least square method. Tendency value represents the linear trend of variables, that is, the rate of increase or decline.

The absolute residuals series f i of y i and the linear fitting series is:
$$f_{i} = \left| {y_{i} - a - bt_{i} } \right|.$$

The \(\bar{f}_{i}\) of the absolute residuals series f i is called the mean fluctuation value.

For the absolute residuals series f i , t i represents the corresponding time, a linear regression equation between f i and t i is as follows:
$$\hat{f}_{i} = c + dt_{i} .$$

Similarly, the regression coefficient d is called the “fluctuation value,” which represents the linear trend of absolute residuals series, that is, the rate of increase or decrease of fluctuation.

2.3 Confidence Test

The result of a confidence test is classified by different indices and confidence standards, shown in Table 1. In the table, “Not significant” means there is no significant tendency or fluctuation under certain confidence.
Table 1

Classified statistics of grids under different confidence standard (%)

Classification of index

Confidence

0–100 %

66–100 %

(Likely)

90–100 %

(Very likely)

99–100 %

(Virtually certain)

Level-I indices

 Temperature tendency

  Rise

97

96

95

93

  Decline

3

2

2

1

  Not significant

 

2

3

6

 Precipitation tendency

  Rise

56

42

33

22

  Decline

44

32

24

15

  Not significant

 

26

43

63

Level-II indices

 Temperature fluctuation

  Increase

46

26

15

6

  Decrease

54

34

23

12

  Not significant

 

40

62

82

 Precipitation fluctuation

  Increase

59

44

34

24

  Decrease

41

25

15

6

  Not significant

 

31

51

70

For level-I indices, 97 % grids are significant with respect to temperature tendency under 90 % confidence, while 57 % grids are significant with respect to precipitation tendency. For level-II indices, grids with significant fluctuation of temperature or precipitation are lower than that of tendency; under 90 % confidence, 38 % grids are significant in respect of temperature fluctuation, while 49 % grids are significant in terms of precipitation fluctuation.

3 Spatial Distribution of Climate Change (1961–2010)

The spatial distribution of temperature and precipitation tendency values is shown in Fig. 2. Temperature tendency shows latitudinal zonality at the world scale and the rising rate is higher in the northern high latitudes than the northern low latitudes and southern hemisphere, while some areas show cooling in the past 50 years on a regional scale, such as the Andes Mountains and northwest Australia. The regional features of precipitation tendency are more complex than those of temperature tendency. The wet trend areas mainly include Europe, Asia, Oceania, North America and Western South America, while the dry trend areas are mainly concentrated in tropical Africa, West Asia, and eastern South America on a continent scale.
Fig. 2

Spatial distribution of temperature and precipitation tendency values. a Temperature tendency from 1961 to 2010, and b precipitation tendency from 1961 to 2010

The spatial distribution of temperature and precipitation fluctuation values is shown in Fig. 3. The areas in which temperature fluctuation is increasing mainly include Western Europe, central Africa, East Asia, Australia, and northern North America, while the areas with decreasing temperature fluctuation mainly cover western and southern Africa, North Asia, Central Asia, southern North America, and South America when examined at the continental scale. For precipitation fluctuation, the increasing areas mainly include Asia and North America, while decreasing areas mainly cover Europe, Africa, West Asia, Australia, and South America.
Fig. 3

Spatial distribution of temperature and precipitation fluctuation values. a Temperature fluctuation from 1961 to 2010, and b precipitation fluctuation from 1961 to 2010

4 World Regionalization of Climate Change (1961–2010)

Based on the regional differentiation patterns of climate change (Sect. 3), we further propose a fundamental method of climate change regionalization that includes a set of basic principles, an indicator system, and a regionalization method. This scheme of climate change regionalization is designed using datasets assembled from the five decades between 1961 and 2010.

4.1 Principles, Indicators, and Method

Five basic principles considered in our regionalization method include:
  1. (1)

    keep intact the boundaries of major administrative subdivisions of countries (the state boundaries of the US, for instance);

     
  2. (2)

    consider only leading climate factors (temperature and precipitation);

     
  3. (3)

    preserve spatial distribution continuity (the essential difference between climate classification and regionalization);

     
  4. (4)

    combine key indices with auxiliary indices (direction of change as key indices, rate of change as auxiliary indices); and

     
  5. (5)

    retain the consistency of large-scale terrain units that are important to land–atmosphere interaction.

     

The regionalization that results from these principles is divided into two levels: climate tendency zones (level-I) and climate fluctuation regions (level-II). For level-I regionalization, we take temperature tendency as the main indicator by which to divide the world into warming-trend zones and cooling-trend zones, while the rate of change is regarded as the auxiliary indicator for determining regionalization boundaries. Precipitation tendency is divided in the same way. In addition, considering that 97 % of the land area showed a warming trend, we further adopted the average rate of temperature rise (0.22 °C/10 a between 1961 and 2010) as the threshold value with which to divide the world into a rapidly rising temperature zone and a slowly rising zone. Specifically, 0.22 °C/10 a is the mean of the temperature tendency values of all the grids located in the global land area except Antarctica (see Fig. 2a).

For level-II regionalization, we take temperature fluctuation as the main indicator that divides the world into temperature fluctuation increasing regions and temperature fluctuation decreasing regions, with the rate of change as the auxiliary indicator for determining regionalization boundary. Precipitation fluctuation is divided in the same way.

The mapping work is implemented in ArcGIS environment, in which the zonal statistics, dissolve, and superimposed tools are adopted. For level-I zones, first apply the zonal statistics tool to calculate the mean value of grids of each unit (administrative subdivisions of countries). Then classify each unit into different types according to the direction of change. Third, use the dissolve tool to combine the units of same type, obtain the zoning result of level-I regionalization. Level-II regionalization is processed with the same method. Finally, apply the superimposed tool to overlay the zoning results of level-I and level-II regionalization and complete the zoning map.

4.2 Result

Temperature and precipitation tendency values (Fig. 2) are adopted to divide the world into six tendency types. The warm(rapid)-wet type accounts for 36.8 % of the global land area, and the warm(rapid)-dry type accounts for 27.6 %, the warm(slow)-wet type accounts for 17.6 %, the warm(slow)-dry type accounts for 15.1 %, the cool-wet type accounts for 1.4 %, and the cool-dry type accounts for 1.5 % of the global land area, respectively.

Level-I regionalization resulted in 12 tendency zones, that is, Europe–North Asia warm(rapid)-wet trend zone, Northern Africa–West Asia warm(rapid)-dry trend zone, Southern Africa warm(slow)-dry trend zone, South Asia–Central Asia warm(slow)-wet trend zone, Northeast Asia–East Asia warm(rapid)-dry trend zone, East Asia–Southeast Asia–Oceania warm(slow)-wet trend zone, Western North America warm(rapid)-dry trend zone, Northern North America warm(rapid)-wet trend zone, Southern North America–Northern South America warm(slow)-wet trend zone, Western South America cool-wet trend zone, Eastern South America warm(slow)-dry trend zone, and Southern South America warm(slow)-wet trend zone. Characteristic values of all climate tendency zones according to classified indicators are shown in Table 2.
Table 2

Characteristic values of level-I world regionalization of climate change (1961–2010)

For warm trend, r: rapid, s: slow

Temperature and precipitation fluctuation values (Fig. 3) are adopted to divide the world into four fluctuation types. The temperature and precipitation fluctuation increasing type accounts for 18.7 %, the temperature fluctuation increasing and precipitation fluctuation decreasing type accounts for 39.9 %, the temperature fluctuation decreasing and precipitation fluctuation increasing type accounts for 19.5 %, and the temperature and precipitation fluctuation decreasing type accounts for 21.9 % of the global land area, respectively.

The level-II regionalization divides the world into 28 fluctuation regions, based on level-I regionalization. Characteristic values of all climate fluctuation regions according to classified indicators are shown in Table 3. Combine level-I regionalization and level-II regionalization, then name them by geomorphic features, the world regionalization of climate change (1961–2010), as shown in Fig. 4, is completed.
Table 3

Characteristic values of level-II world regionalization of climate change (1961–2010)

T.F.: temperature fluctuation, P.F.: precipitation fluctuation

Fig. 4

World regionalization of climate change (1961–2010)

5 Discussion

In addition to the calculation reported in this article, we have also applied an ensemble empirical mode decomposition method and Bayesian approach to analyze the regional differences of climate indicators, and the patterns are approximately the same. However, it is worth mentioning that there are different understandings concerning the variation change. Considering the study period of the last half-century, in this article we focused on interdecadal changes. In the future, we will try to diagnose climate change on interannual, seasonal, and monthly scales, which should be conducive to analysis of the relationship between climate change and environmental risks.

There are essential differences between climate change regionalization and traditional climate regionalization. The result of climate change regionalization depends on the spatial and temporal scales selected, that is, climate change regionalization shows certain differences at various scales. Next step, we shall carry out research also at different spatial scales to further reveal the variation range of climate change patterns.

Considering that existing disaster databases of the world mainly use countries or regions as the basic spatial unit and lack spatial specificity, this study selected administrative subdivisions of countries as the basic unit of the regionalization. Our research aimed to probe an important impacting factor in disaster risks and to reveal the driving forces of climate-related disasters and environmental risk of countries in the world through understanding the regional differences of climate change.

6 Conclusion

This article is an outgrowth of earlier work (Shi et al. 2014), which explores the pace and direction of climate change in China over the last half century. Our primary goal was to develop a methodology and model that could generate a map that accurately reflect a scientifically defensible understanding of regional variation in climate change. In the current article, we have extended this model to a global scale and have demonstrated that it is technically possible to create a spatially nuanced approximation of climate change. With five decades of data at our disposal, it is possible to separate variation from trend and identify subnational areas impacted by climate change. In this sense, our work contributes a technical platform to the study of regional climate change. Our method potentially can be used to explore the scale, direction, and pace of climate change in a variety of spatial and temporal contexts. The model also offers the prospect of positive employment in disaster risk management and planning. By identifying regions most threatened by rapid, climate change-induced instability, disasters may be better mitigated and risks reduced.

Footnotes

References

  1. Dai, A., I.Y. Fung, and A.D. Del Genio. 1997. Surface observed global land precipitation variation during 1900–88. Journal of Climate 10: 2943–2962.CrossRefGoogle Scholar
  2. Emori, S., and S.J. Brown. 2005. Dynamic and thermodynamic changes in mean and extreme precipitation under changed climate. Geophysical Research Letters 32(17): L17706. doi: 10.1029/2005GL023272.CrossRefGoogle Scholar
  3. Hansen, J., R. Ruedy, M. Sato, and K. Lo. 2010. Global surface temperature change. Reviews of Geophysics 48(4): RG4004. doi: 10.1029/2010RG000345.CrossRefGoogle Scholar
  4. IPCC (Intergovernmental Panel on Climate Change). 2012. IPCC special report on managing the risks of extreme events and disasters to advance climate change adaptation (SREX). Cambridge: Cambridge University Press.Google Scholar
  5. IPCC (Intergovernmental Panel on Climate Change). 2013. IPCC climate change 2013: The physical science basis. Cambridge: Cambridge University Press.Google Scholar
  6. Ji, F., Z.H. Wu, J.P. Huang, and E.P. Chassignet. 2014. Evolution of land surface air temperature trend. Nature Climate Change 4: 462–466.CrossRefGoogle Scholar
  7. Köppen, W.P. 1936. The geographical system of climates (Das geographische system der klimate). In Handbook of climatology (Handbuch der klimatologie), ed. W.P. Köppen, and R. Geiger, 1–44. Berlin: Gebrüder Borntraeger.Google Scholar
  8. Mitchell, T.D., and P.D. Jones. 2005. An improved method of constructing a database of monthly climate observations and associated high-resolution grids. International Journal of Climatology 25(6): 693–712.CrossRefGoogle Scholar
  9. Schneider, U., A. Becker, P. Finger, A. Meyer-Christoffer, B. Rudolf, and M. Ziese. 2011. GPCC full data reanalysis version 6.0 at 0.5: Monthly land-surface precipitation from rain-gauges built on GTS-based and historic data. doi: 10.5676/DWD_GPCC.
  10. Shi, P.J., S. Sun, M. Wang, N. Li, J.A. Wang, Y.Y. Jin, X.T. Gu, and W.X. Yin. 2014. Climate change regionalization in China (1961–2010). Science China Earth Sciences 57(11): 2690–2701.CrossRefGoogle Scholar
  11. Strahler, A.N. 1975. Classification of global climates. In Physical geography, ed. A.N. Strahler, 243–245. New York: John Wiley.Google Scholar
  12. Thornthwaite, C.W. 1948. An approach toward a rational classification of climate. Geographical Review 38(1): 55–94.CrossRefGoogle Scholar
  13. Wu, Z.H., N.E. Huang, J.M. Wallace, B.V. Smoliak, and X.Y. Chen. 2011. On the time-varying trend in global-mean surface temperature. Climate Dynamics 37: 759–773.CrossRefGoogle Scholar

Copyright information

© The Author(s) 2016

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  • Peijun Shi
    • 1
    • 2
    • 3
  • Shao Sun
    • 1
    • 2
    • 3
  • Daoyi Gong
    • 1
    • 3
  • Tao Zhou
    • 1
    • 3
  1. 1.State Key Laboratory of Earth Surface Processes and Resource EcologyBeijing Normal UniversityBeijingChina
  2. 2.Key Laboratory of Environmental Change and Natural Disasters of Ministry of EducationBeijing Normal UniversityBeijingChina
  3. 3.Academy of Disaster Reduction and Emergency ManagementMinistry of Civil Affairs & Ministry of EducationBeijingChina

Personalised recommendations