Climatic regionalization of Serbia was performed in the paper on the basis of data on air temperature from 23 meteorological stations and using cluster analysis. Applying Mann-Kendall test and Sen's slope estimates, temperature linear trends were investigated within each of the clusters, but also at each station individually. It has been shown that a positive linear trend is present in practically entire territory of Serbia and that it is very strong in Belgrade. Differences in air temperature between Belgrade and the cluster in which it is located were examined on an annual basis. [Projekat Ministarstva nauke Republike Srbije, br. III 47007]

In this study, the projected changes of the average monthly and annual air temperatures, precipitation amounts and De Martonne aridity index in Serbia in the 21st century are examined. For that purpose, the spatial distribution of air temperature and precipitation amount is statistically modelled using data of 64 climatologic stations and 421 precipitation stations for the period 1976–2005 using multiple linear regression. The De Martonne aridity index is calculated from the spatially distributed data. In order to obtain regional averages and variability, we divide Serbia into three subdomains: the western zone (the Dinarides), eastern mountainous (Carpathian‐Balkan) zone and the terrains with altitudes lower than 500 m a.s.l. (Lowland). As a second step, the members of the COordinated Regional Downscaling EXperiment (CORDEX) for Europe (EURO‐CORDEX) are evaluated in terms of near‐surface air temperature and precipitation for Serbia. By comparing respective values for the period 1976–2005 with future periods 2021–2050 and 2071–2100, we derive the climate change signal for near‐surface air temperature, precipitation and the De Martonne index for both scenarios RCP4.5 and RCP8.5. Hereby, the ensemble averages are calculated as weighted averages using the inverse standardized Euclidean distance by comparing observation and model results for near‐surface air temperature and precipitation as a weight. We find for the period 2071–2050 that most ensemble members predict a statistically significant increase in monthly and annual near‐surface air temperature in both scenarios in all three subdomains, and a statistically significant decrease in the yearly De Martonne index in the RCP8.5 scenario in the Carpathian‐Balkan and Dinarides. For the last decades of the 21st century, our findings indicate much higher temperatures and increased risks of drought for the RCP 8.5 scenario compared to both the observations from 1976 to 2005 and the RCP 4.5 scenario.

The impact of solar activity on environmental processes is difficult to understand and complex for empirical modeling. This study aimed to establish forecast models of the meteorological conditions in the forest fire areas based on the solar activity parameters applying the neural networks approach. During July and August 2018, severe forest fires simultaneously occurred in the State of California (USA), Portugal, and Greece. Air temperature and humidity data together with solar parameters (integral flux of solar protons, differential electron flux and proton flux, solar wind plasma parameters, and solar radio flux at 10.7 cm data) were used in long short-term memory (LSTM) recurrent neural network ensembles. It is found that solar activity mostly affects the humidity for two stations in California and Portugal (an increase in the integral flux of solar protons of > 30 MeV by 10% increases the humidity by 3.25%, 1.65%, and 1.57%, respectively). Furthermore, an increase in air temperature of 10% increases the humidity by 2.55%, 2.01%, and 0.26%, respectively. It is shown that temperature is less sensitive to changes in solar parameters but depends on previous conditions (previous increase of 10% increases the current temperature by 0.75%, 0.34%, and 0.33%, respectively). Humidity in Greece is mostly impacted by solar flux F10.7 cm and previous values of humidity. An increase in these factors by 10% will lead to a decrease in the humidity of 3.89% or an increase of 1.31%, while air temperature mostly depends on ion temperature. If this factor increases by 10%, it will lead to air temperature rising by 0.42%.

The aim of this paper is to examine the spatial and temporal variability of the average monthly, seasonal and annual air temperatures in Serbia. Therefore, data from 64 climatologic stations were analyzed in the period from 1961 to 2010. Based on the data, on the position of the stations (their latitude, longitude, altitude), and the characteristics of the terrain in their vicinity (inclination and terrain exposure in a radius of 10 km around the station), a regression model was constructed based on which air temperatures are interpolated for the territory of Serbia. The root-mean-square error (RMSE) of the regression model ranged from 0.2 ºC in January, February and November to 1.1ºC in August. Spatial distribution of air temperatures is shown (maps of mean monthly, mean seasonal and mean annual air temperatures are made), and the Sen's procedure was used to calculate trends of air temperatures (maps of average monthly, mean seasonal and mean annual trends of air temperatures). The Mann-Kendall test was used to test the significance of air temperature trends. Apart from the southeast, the whole territory of Serbia has practically experienced a statistically significant rise in the average annual air temperature, with the highest increase in the summer and winter months. [Project of the Serbian Ministry of Education, Science and Technological Development, Grant no. III47007]

Increasing thermal risk in cities is endangering the health and well-being of urban population and is driven by climate change and intensive urbanization. Therefore, if we plan to enlarge the capacities of cities to be more climate resilient in the 21st century, more detailed monitoring of urban climate on local and micro scales is needed. For this research we performed two microclimate measurement campaigns in urban area of Belgrade, during hot summer days in 2021. In total, five measurement sites were chosen in different urban designs and different local climate zones (LCZs). For thermal monitoring (air temperature – Ta and globe temperature – Tg) the Kestrel heat stress tracker sensor with 1-min measurement resolution was used, but we used 10-min average values. Obtained results showed distinct thermal differences (up to 7 °C on average) between densely built-up areas and green areas. Differences between built-up LCZs are lower with values from 2 to 4 °C. Important part of this research was microclimate monitoring on sites within the same LCZ (intra-LCZ variability). Results showed that shadows and short- and longwave radiation play a paramount role in thermal variability. Direct and reflected radiations on one measurement site increased Ta up to 6 °C and Tg up to 12 °C when compared to other measurement site (in a similar urban design), which was in the shadow.