Future Scenario of Global Climate Map change according to the Köppen-Geiger Climate Classification

Earth’s climate changes rapidly due to the increases in human demands and rapid economic growth. These changes will affect the entire biosphere, mostly in negative ways. Predicting future changes will put us in a better position to minimize their catastrophic effects and to understand how humans can cope with the new changes beforehand. In this research, previous global climate data set observations from 1961-1990 have been used to predict the future climate change scenario for 2010-2039. The data were processed with Idrisi Andes software and the final Köppen-Geiger map was created with ArcGIS software. Based on Köppen climate classification, it was found that areas of Equator, Arid Steppes, and Snow will decrease by 3.9 %, 2.96%, and 0.09%, respectively. While the areas of Warm Temperature and Dessert will increase by 4.5% and 0.75%, respectively. The results of this study provide useful information on future climate KöppenGeiger maps and areas that will most likely be affected by climate change in the following decades.


Introduction:
Global climate classifications were developed to distinguish the climates into a number of types (1) based on climate's individual elements, such as precipitation, temperature, wind, vegetation distribution, and atmospheric pressure (2,3). Climate classifications are established to display the mean spatial climate characteristics; also, it is used to identify climatic types (4). There are two classification systems: the empirical that uses statistical data and the genetic that uses causal factors to classify. The most used method for climate classification (5) is the Köppen Climate Classification developed in 1900 (5) by the plant physiologist Wladimir Köppen. Natural vegetation is used by Köppen as an index for different climates (6). Köppen suggested the first scheme in the description of climatic types based on numerical limits of temperature and precipitation that represented the ecological margin of similar flora (2). Five basic climate types have been distinguished (7). These types are presented via the first five letters of the alphabet: (A) for the equatorial zone, (B) for the arid zone, (C) for the warm zone, (D) and (E) t for the snow zone and polar zone respectively. The climate subtypes are symbolized by a second or a third letter, depending on which parameter is taken into consideration for the classification (6). The global mean surface temperature has risen by about 0.5 °C over the last 100 years (8), and global average surface air temperature, which is affected by land cover, vegetation phenology, anthropogenic effects, and geology (9) is projected to increase climate change in different regions of the globe in a different way (10) This classification is based on two parameters: the annual cycles of precipitation and temperature (7). Surface air temperature is the most commonly used climate change parameter (11), and together with precipitation are the two elements that affect our environment more noticeably and directly lead to reduce lakes ice cover in the world (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12). This method also takes into concern the potential of  (3)(4)(5)(6) and applied it to confirm general circulation control model runs of present climate and greenhouse gas warming simulations.
In the current research, a future scenario generated in order to produce a future map based on SRES (Special Report on Emissions) A1B emission scenario (14-15) that assume: "a future world with a fast-economic growth, low population growth and the expiation of using new and more advanced technology. However, the major underlying themes are economic, cultural convergence, and capacity building, with a reducing regional difference in per capita income. In a world where people work for individual wealth rather than the quality of environmental." In addition, the past decade available data used for the creation of a second world map, including real measurements and a high-resolution grid.
The work aimed to create two climate zones maps for past and future based on available temperature and precipitation data for the thirty years  in the past and thirty years (2010 -2039) future predicted data based on SRES A1B scenario, then compare and calculate the area of the climate zones.

Data and Methods
Two global datasets of climate observations regarding the past data and the future scenarioproviding climate modeled datasets have been selected for this research. All the data are obtained from the Intergovernmental Panel on Climate Change (IPCC) are available on www.ipccdata.org/. Monthly precipitation and temperature observations for the past 30 years from the Climate Research Unit, CRU of the University of East Anglia, are obtainable on a 0.5-degree latitude/longitude grid resolution and cover the global land areas excluding Antarctica. These data cover a period of 30 years, from 1961 to 1990. Future precipitation and temperature scenario data (SRES A1B Emissions Scenarios) are available on a 2.81 degrees' latitude/longitude grid, with monthly resolution and excluding Antarctica. These data cover 30 years, from 2010 to 2039. All data above have been processed with the GIS analytical tool land change modeler IDRISI Andes software (version 15.00). The software provides a rapid quantitative assessment of change by graphing gains and losses by land cover categories, and the final maps were created and processed in Geographic information system ArcGIS software (version 10.8). The calculation of Köppen classes' classification was as follows, the main climate classes, represented by one or two letters depending on the climate differences described in Table 1.
In Table 1 (Köppen classification), the first letter represents the main class and the second one, the subclass. These summed seven different possible climate classes that are used in this study. The classes A, C, and D are not divided into any subclasses; therefore, they will not be considered. Classes B and E have been divided into the subclasses BS (Steppe climate), BW (Desert climate), ET (Tundra climate) and EF (Frost climate). Köppen and Geiger observed that in a large areal amount, not all of the subclasses occur, and not all the class types had the same climatological importance (9); accordingly, in the present article, just seven classes have been taken into account. In Table 1 (Köppen classification), the first letter represents the main class and the second one, the subclass. These summed seven different possible climate classes that are used in this study. Climatology was calculated by adding separately all years' months and dividing them by 30. For example, the climatology of the temperature of January (for the past) should be calculated as (Jan. "61 + Jan. "62 + …+ Jan. "90)/30, which finally produces the mean -30 yearstemperature value for the selected month. the monthly mean temperature of the warmest month, and T min the monthly mean temperature of the coldest month. All temperatures are given in degree Celsius (°C), precipitation in cm/month, and annual precipitation in cm/year. Additionally, P th (dryness threshold) was calculated in cm regarding the arid climates (B class), which depend on absolute | Tann |, The absolute value of the mean annual temperature, and on the yearly period of precipitation (9). After completing those calculations, the classes and subclasses A, C, D, EF, and ET have been created, based on the criteria from Table 1. Antarctica is not included within the data sets, and therefore, this region is also excluded from the analyses presented here, but it is reported as a polar frost climate (EF) in the (9) study.
For the creation of the arid class (B) and its subclasses (BW and BS), the criteria from Table 1 and equation (Eq.1) were used. The global data divided into two hemispheres, separating the South and North winter and summer respectively, and afterward, the hemispheres merged, based on the correct season.
All the separate maps created from the previous steps were finally added together, composing the final global past and future maps, these presented in the results. 2{T ann } + 28 if at least 2 3 ⁄ of the annual precipitation occurs in the summer. Otherwise.

Results:
Two different sets of data (resolution) used 2.81 degrees (approximately more than 300 Km 2 for each pixel size) and 0.5 degrees (approximately more than 55 Km 2 for each pixel size) latitude/longitude grid in this research; this is because the data for future scenario (A1B SERS) was only available on a regular 2.81-degree latitude/longitude grid. Figure 1 presents a world map produced from the mean monthly temperature and precipitation data according to Köppen climate classification, for the period 1961 to 1990 on a regular 0.5-degree latitude/longitude grid later it was changed to 2.81degree resolution Figure 2 as in future data Figure 3 in order to compare and calculate the accurate area for climate zones Figures 4 and 5, the future data set were only available in 2.81 degrees. Figures 2 and 3 show the world maps resulting from the Köppen-Geiger climate classification using mean monthly temperature and precipitation data for the periods 1961 to 1990 and 2010-2039, respectively, on a regular 2.81degree latitude/longitude grid. All seven-climate classes are illustrated with different colors. Data for Antarctica were not available, and the Antarctic region has been set manually to the ocean. The resulting world maps in Figs. 2 and 3 for 30 years past and future respectively, show how the classes" boundary is and how the area of each class changes, the climate prediction for the future is based on past data and general trends of temperature and precipitation change.
In the graph presented in Fig. 4, the area (km 2 ) of each class in the past and the future are reported. In some cases, like A, C, and BS class, the difference is significant, and in some others, such as D and ET class, no significant alteration was found. The calculated percentages show that A-class will decrease by 3.9% as well as the BS class by 2.96%. A slight reduction could predict for the D class by 0.09%. Class C will have a more considerable increase by 4.5, while classes ET, BW, and EF are increased by 0.15%, 0.75%, and 0.85% respectively as shown in Fig. 5.

Discussion:
Sparovek, et al. (5) suggested that specific tools of Geographical Information System (GIS) software can improve the accuracy and quality of the maps created for the Köppen climate classification and; it also used to intraplate more parameter like river pattern with Köppen-Geiger climate zone for creating new climate classification like hydroclimate zones (16). Köppen Climate Classification is considered a very old method for climate classification but still is the most valuable method (6), and it is more likely to be used for another century (11).   In the current article, Köppen Climate Classification is used to create a detailed world climate map, based on observed and predicted data sets, for the past and future 30 years. Previous studies, such as (8), (17) and (18) have used this classification to produce corresponding updated world maps. Furthermore, other researchers have used past data, modeled future data and paleoclimate data to observe Arctic climate alterations (19,20). In this article, the period 1961-1990 has been used as the "past 30 years' period" for the climate classification-model. When the past 30 years are mentioned in the article, the gap of 20 years should be considered. Not until recently, there have been other updates of this climate classification-model; even though the data is from the past three decades, it is still considered to emulate the present reasonable period (6). Past and expected future scenario's temperature and precipitation data have been used to create the seven climate classes and subclasses of the classification. The additional four subclasses (BS, BW, ET, and ET) that are the polar and arid subclasses were chosen to be calculated, so a comparison between the two threshold climates of the earth could be possible. This comparison revealed that the areas of Equatorial (A), Arid Steppe (BS) and also Snow (D) climates will decrease and, in some cases, this could be explained by the increase of the temperature (19,21). On the other hand, areas of Warm Temperature (C), Dessert Polar Frost (EF) climate might be a software error. In comparison with other similar studies about climate change and global warming (19) suggested that this error could be as a software, a user, or a dataset error.

Conclusion
A climate change prediction is indispensable for the development of robust climate adaptation plans. The results of this study provide useful information on future climate Köppen -Geiger maps and areas that will most likely be affected by climate change in the following decades. The dessert area will increase by 4.5 %, while the equator and arid steppes zone will decrease by 3.9 % and 2.96%, respectively. These changes will directly affect human wellbeing by altering the availability of freshwater, land-use, and food production across the world.