Climate Change Is Leading to a Convergence of Global Climate Distribution

The impact of changes in global temperatures and precipitation on climate distribution remains unclear. Taking the annual global average temperatures and precipitation as the origin, this study determined the climate distribution with the distances of temperature and precipitation from their global averages as the X and Y axes. The results showed that during 1980–2019, the global temperature distribution converged toward the mean (convergence), while the precipitation distribution moved away from the mean (divergence). The combined effects of both led to a convergence in the global climate distribution. During 2025–2100, significant climate convergence is observed under two emission scenarios (SSP245 and SSP585). However, the climate convergence and the area of change in climate type remains insignificant only under SSP126, suggesting that the diversity of the global climate pattern can be maintained under a sustainable emission pathway (SSP126), whereas high emission pathways will lead to greater uniformity in global climate.


Introduction
Global climate change determines land vegetation distribution (Gerten et al., 2004;Reyer et al., 2013), impacts species diversity (Blois et al., 2013;Pecl et al., 2017;Rinawati et al., 2013), has significant effects on vegetation productivity (Cao et al., 2021), and poses a threat to global food security (Cheruiyot et al., 2022;Fones et al., 2020;Richards et al., 2021).According to the Sixth Assessment Report of the United Nations Intergovernmental Panel on Climate Change (IPCC6) human activities have led to a global warming of approximately 1.3°C above pre-industrial levels, and if the current rate of warming continues, global temperatures could reach 1.5°C between 2030 and 2052.Global warming exhibits regional disparities, with temperatures in the Eurasian continent increasing 2 to 3 times faster than the global average, and the Arctic and Antarctic Peninsula experiencing higher increases by 3 to 4 times (Hansen et al., 2010).Furthermore, the global water cycle is accelerating (Cui et al., 2018), resulting in increased annual precipitation globally, but with highly prominent regional variations.Since the 21st century, precipitation has decreased in the north of the equatorial Pacific and in southern Africa, whereas it has increased in the south of the equator and increased in the mid to high latitudes (Thorpe & Andrews, 2014).Therefore, climate warming and changes in the global water cycle have significant impacts on the global climate distribution (Domínguez-Tuda & Gutiérrez-Jurado, 2021;Dore, 2005;Ezaz et al., 2022).
The distribution of climate is a comprehensive reflection of temperature and precipitation (Wang et al., 2010).Currently, climate distribution has been defined on the basis of various factors such as temperature, precipitation, latitude zones, and underlying surfaces.Examples of the definition include the Köppen climate classification and global climate classification.Climate change has led to an increase in the frequency and intensity of extreme high temperatures (including heatwaves) in most terrestrial regions worldwide, along with an increase in the frequency and intensity of heavy precipitation events.Current findings suggest that the global climate is becoming increasingly extreme.However, is this impression truly accurate?
To answer this question, this study established a climate Cartesian coordinate system.In this system, the annual global averages of temperature and precipitation are taken as the origin, with temperature on the x-axis and precipitation on the y-axis.For each grid cell, differences in temperature and precipitation from the global averages are represented as x and y, respectively, while the Euclidean distance between a grid cell and the origin is denoted as d.These x, y, and d values represent the deviation of temperature, precipitation, and climate from global averages.Therefore, these three indicators can reflect the homogeneity of the global climate distribution, which is herein referred to as climatic convergence.
Based on this coordinate system, the first, second, third, and fourth quadrants are defined as four climate types: warm-wet, cold-wet, cold-dry, and warm-dry (Ahlström et al., 2015;Batlle-Bayer et al., 2014;Li et al., 2021;Salehnia et al., 2017).Specifically, this study explored whether the global climate distribution has become more extreme or more convergent over the past 40 years and how different emission scenarios will affect the climate distribution in the future.The findings offer a novel perspective on global climate change.

Climate Data
The temperature and precipitation data provided by CRU for the period from 1980 to 2019 have a spatial resolution of 0.5°.For the period 2025-2100, three Shared Socioeconomic Pathways (SSPs) were selected: SSP1-2.6,SSP2-4.5, and SSP5-8.5.Temperature and precipitation data under these three pathways were obtained from the ACCESS-CM2 and NorESM2-LM models, which were selected considering their low error and good agreement with historical data (Du et al., 2022;Zhu & Yang, 2020).These model data have a spatial resolution of 2.5°.Data preprocessing was conducted using ArcGIS 10.4, with a unified coordinate system of WGS84 (Plate Carree (world)).Annual mean temperature, annual total precipitation, and area statistics were projected using the Goodes homolosine projection.

Calculation of Climate Convergence
The data were first standardized using Z-score normalization method.The definition of convergence is as follows: First, establish a coordinate system.In the coordinate system, the origin is set at the annual average temperature and precipitation.The x and y axes represent the distance of each grid's temperature and precipitation from the global average values for each year.Simultaneously, we define this distance (C tem and C pre ) as the temperature and precipitation convergence of each grid, while the Euclidean distance indicates the climate convergence (C cli ).Based on the signs of C tem and C pre , the climate type of each grid can be determined, and the coordinate system is illustrated in Figure 1. (1) C tem and C pre can be positive or negative, with smaller absolute values indicating closer convergence to the mean.A smaller C cli , greater than 0, also suggests closer convergence to the mean.Global convergence is determined by averaging all grid convergence values.

Distribution and Variation of Global Climate Convergence
During the period from 1980 to 2019, both global average temperature and total precipitation exhibited fluctuating upward trends.Both temperature and precipitation exhibited an increasing trend in more than half of the world's regions, accounting for 98.15% and 61.26% of the global area, respectively.In particular, temperature and precipitation showed statistically significant trends (P < 0.05) in 85.60% and 14.53% of these regions, respectively (Figure S1 in Supporting Information S1).
Temperature convergence exhibited a clear latitudinal distribution, with the mean temperature zone (where convergence is near 0) situated near the 45°l atitude lines, as indicated by the blue region in Figure 2a.Within the latitudes between the northern and southern 45°lines, temperature convergence was greater than 0 (depicted in red), and convergence increased with increasing latitude, reaching its maximum near the equator.In regions with latitudes greater than 45°, temperature convergence was less than 0 (depicted in green), and the absolute value of convergence also increased with increasing latitude, with the highest convergence observed in the polar regions (Figure 2a).The mean zone for precipitation convergence (indicated in blue) was less widespread globally.Representing convergence greater than 0, the red and yellow areas were primarily concentrated in regions with abundant precipitation, such as the tropics, eastern North America, South Asia, and East Asia.The green areas representing convergence less than 0 were mainly distributed in subtropical and high-latitude regions, where precipitation was lower than the global average (Figure 2b).
Climate convergence also exhibited a clear latitudinal distribution.The primary region for the global climate mean, where convergence is near zero, was situated around the 45°latitude in both the northern and southern hemispheres.This zone extends from the southern United States in North America, across central and western  Europe, to the eastern parts of Asia.There are also smaller areas with similar characteristics in western and southern South America, the southern tip of Africa, and the southern tip of Australia.With distance from the mean zone toward either the north or south, climate convergence gradually increased.It peaked near the equator and then gradually decreased again with proximity toward the poles (Figure 2c).Regions with the most extreme climate convergence were primarily located in the Arctic, followed by areas near the equator.During the period from 1980 to 2019, both global climate convergence and temperature convergence exhibited a decreasing trend.In contrast, precipitation convergence showed an increasing trend (Figure 2d).This indicates that global temperature distribution exhibited a pronounced trend of significant convergence over the past 40 years, while precipitation distribution showed less significant divergence.Collectively, these trends have led to a convergence in climate distribution.The variations in temperature, precipitation, and climate convergence among the four climate types also generally follow the global trends (Figure S2 in Supporting Information S1).
On the whole, global temperature has been displaying a trend toward convergence (Figure 2d).Regions exhibiting convergence accounted for 69.47% of the total, whereas regions with diversity accounted for 30.53% (Figure 3a).In contrast, the distribution of precipitation exhibited an overall trend toward diversity (Figure 2d).Areas with diverse precipitation covered 59.88% of the total, whereas convergent areas accounted for 40.12% (Figure 3b).In terms of global climate distribution, an overall trend toward convergence was observed, with convergent regions accounting for 54.58% of the total, including 16.27% with P < 0.05.Areas with convergent climate were primarily distributed across Eurasia, Greenland, South America, Africa, and Australia, among others (Figure 3c).Conversely, regions exhibiting climate diversity accounted for 45.42% of the total.Climate divergence was primarily driven by precipitation distribution, and areas with diverse climate were concentrated around Eurasia, including central Eurasia, North Africa, the Arabian Peninsula, and Iran.They also extended to a significant portion of southern North America.

Simulation of Climate Distribution Under Future Scenarios
During the timeframe spanning from 2025 to 2100, under the SSP1-2.6 scenario, there is no significant trend toward climate convergence.Conversely, under the SSP2-4.5 and SSP5-8.5 scenarios, there is a notable and statistically significant trend toward global climate convergence (Figures 4b, 4d, and 4f).Temperature distribution exhibits a highly significant convergence trend under all three scenarios, but precipitation does not exhibit a significant divergence trend under the SSP1-2.6 scenario.Under the SSP2-4.5 and SSP5-8.5 scenarios, precipitation does not exhibit significant convergence.The spatial distribution of climate convergence shows notable differences among the three scenarios.Under the SSP1-2.6 scenario, the area of climate convergence accounts for 58.30%, including 15.51% with P < 0.05, and the area of divergence accounts for 41.70% (Figure 4 a).Under the SSP2-4.5 and SSP5-8.5 scenarios, convergence accounts for 65.90% and 63.66% of the area, including 27.52% and 37.24% with P < 0.05, and divergence accounts for 34.1% and 36.34%,respectively (Figures 4c and 4e).Under all three scenarios, both the area of significant convergence and the convergence rate gradually increased over time.
During the period from 2025 to 2100, under the SSP1-2.6 scenario, the areas of the four climate types do not show significant changes, consistent with the patterns observed in the area changes from 1980 to 2019 (Figures S3b and  S3d in Supporting Information S1).However, under the SSP2-4.5 and SSP5-8.5 scenarios, the areas of three climate types (excluding the warm-dry climate type) show significant changes.Under the SSP1-2.6 scenario, the differences in the areas of the four climate types remain relatively stable.However, under the SSP2-4.5 and SSP5-8.5 scenarios, the areas of cold-dry and warm-wet climates show significant decreases, whereas the areas of warm-dry and cold-wet climates increase.Overall, the differences in the areas of these four climate types will decrease (Figures S3d,S3f,and S3h in Supporting Information S1).In terms of distribution, the cold-dry areas in the Eurasian continent are decreasing, while the cold-wet areas are expanding, particularly evident under the future SSP5-8.5 scenario.In northern Australia, the warm-dry climate is being replaced by the warm-wet climate (supplementary material Figures S3a, S3c, S3e, and S3f in Supporting Information S1).

Convergence of Global Climate Distribution Pattern
Between 1980 and 2019, approximately 85.60% of global regions exhibited a significant warming trend (p < 0.05), with only 1.85% experiencing cooling.This phenomenon is attributable to the influence of greenhouse gases (CO2, CH4, and N2O) emitted through human activities (Masson-Delmotte et al., 2018).Precipitation showed an increasing trend in approximately 61.26% of the areas, with more significant changes observed near the equator and in high-latitude regions of the Northern Hemisphere (Bradley et al., 1987;Huber & Gulledge, 2011).Conversely, precipitation showed a decreasing trend in approximately 38.74% of regions, aligning closely with the findings of Sun et al. (2014).
In regions with below-average temperatures, particularly in the high latitudes of the Northern Hemisphere, the rate of warming has been higher, while areas near the equator and regions with higher temperatures have experienced slower warming (Hu et al., 2021;Pepin & Lundquist, 2008;Sanderson et al., 2011;M. Serreze et al., 2009;M. C. Serreze et al., 2011).Consequently, differences in global temperature have narrowed, indicating a trend toward approaching the global mean temperature.Furthermore, changes in precipitation exhibited a pronounced spatial variation.Dry regions such as Iran have seen a decrease in precipitation, while wetter regions such as the South Asian monsoon zone have experienced increased rainfall.Consequently, the spatial distribution of climate is characterized by "dry getting drier and wet getting wetter" (Dore, 2005;Kumar et al., 2015;Liu & Allan, 2013).The distribution of precipitation is moving toward diversification, deviating further from the global mean.
The convergence of global climate distribution is notably driven by temperature changes mostly.This is primarily attributed to the increased energy in the atmosphere due to global warming, which accelerates atmospheric exchanges and circulation, leading to a homogenizing effect.Extreme weather events also play a significant role in driving convergence, as they represent the release of accumulated energy over large areas, acting as engines for mixing.Comparing global climate distribution before and after extreme weather events, convergence phenomena can be consistently observed (Li et al., 2022;Lubchenco & Karl, 2012).As shown in Figure 2d, global climate convergence has decreased following extreme weather events.

Distinct Global Climate Type Disparities Under Different Climate Scenarios
Under SSP1-2.6,changes in the areas of the four climate types are not statistically significant, whereas under SSP2-4.5 and SSP5-8.5, the differences in the areas of these climate types significantly decrease.Particularly under SSP5-8.5,around the year 2100, the area covered by warm-dry climates surpasses that under warm-wet climates, indicating a substantial impact of different emission scenarios on the distribution of climates.From 2025 to 2100, climate distribution exhibits significant convergence under SSP1-2.6,SSP2-4.5, and SSP5-8.5, with distinct spatial variations in the trend.SSP1-2.6 represents a path of sustainable socio-economic development, aiming to maintain warming below 2°C, while 2-4.5 and 5-8.5 represent intermediate and fossil-fueldominant development paths, leading to rapid warming.These differences in climate distribution changes across the three scenarios are a direct result of their unique emission characteristics.
Climate distribution convergence implies that climates worldwide are moving closer to the global average, and many developed regions have been observed to be situated within the global climate mean area (Figure 2c).The expansion of this climate mean area implies more favorable conditions for development.However, according to the second law of thermodynamics and the principle of entropy increase, systems tend to evolve spontaneously toward equilibrium, which is a natural tendency from order to disorder and a form of decline.Therefore, convergence may reflect the transitioning of the climate from ordered zonal distribution to random evolution, representing a form of decline.Hence, the issue of the more suitable scenario for humanity remains a topic worthy of further exploration.

Conclusions
As global warming intensifies, temperature distribution tends to converge toward the global average, while precipitation distribution tends to diverge from it.Under the combined influence of these factors, the climate distribution exhibits a tendency of convergence toward the global average.Moreover, differences in the areas under the four climate types decrease.In essence, climate change is driving the global climate distribution toward a state of greater homogeneity.It is worth noting that the impact on climate distribution is highly variable depending on the emission pathway considered.Under a 2°C control target, the global climate distribution would largely retain its current configuration.However, if this target is exceeded, the world would undergo accelerated transformation into a more uniform and homogeneous climate distribution with fewer regional variations.
Is a global climate with diverse regional distribution more preferable, or a more uniform one desired?To maintain the former, carbon emissions should be reduced to control temperature increases.Nevertheless, the latter climate distribution with fewer regional differences may offer more room for development, which may also be desirable.At present, most discussions revolve around controlling temperature, but actual future conditions may be overlooked if existing temperature control goals are not achieved.Nevertheless, possible advantages of a uniform global climate distribution can be further explored.It is important to note that this study did not delve into the underlying mechanisms of climate convergence from an atmospheric perspective, which is a limitation of our research.

Figure 1 .
Figure 1.Climate convergence index coordinate system.Note: The closer to the origin (black arrow), the lighter the color, indicating convergence, and the darker the color is, the farther away from the origin (white arrow), indicating divergence.

Figure 2 .
Figure 2. Spatial distribution of convergence degree of temperature (a), precipitation (b), climate convergence distribution (c), and global average convergence degrees over 1980-2019 (d).Note: Figures(a-c) legends are the convergence degree classification (area ratio).

Figure 3 .
Figure 3. Distribution and trend of temperature convergence degree (a), precipitation convergence degree (b) and climate convergence (c) during 1980-2019.