Begegnungen
Schriftenreihe des Europa Institutes Budapest, Band 28:43–58.

JÁNOS MIKA – ÁKOS NÉMETH

Climatic Characteristics and Tendencies of Climate in Bulgaria and Romania

The European Union is to be enlarged soon by two new South-Eastern European countries, Bulgaria and Romania. The climate of these two countries has certain characteristics, the knowledge of which may promote their integration, and also the preparation of other countries, including our own, for being able to develop advantageous and prosperous relations with these states after their accession.

Both countries are characterised by a highly variegated topography, asserting not only the specificities of mountainous climate, or dividing it from a given area. The semicircle of the Carpathian Mountains in Romania and the longitudinal range of the Rodope in the case of Bulgaria have a clear dividing role in slowing down the air masses coming from behind the mountains, from north and west respectively, and forcing them to rise. In other words, these two mountain ranges strongly modify the image otherwise shaped by the general circulation of air in the region, which is manifest in several components of the climate.

Annual Cycle of the Climate and Its Regional Specificities

At first the characteristic annual schedule of temperature and precipitation is presented (Figures 1–2). In both cases the results are restricted to the capital cities, both of which represent the climate of flatlands. In the next two Figures we would return to the territorial differences. On Figures 1 and 2 the data of Budapest are also given besides those of Bucharest and Sofia, thus offering possibilities of comparison concerning the annual progress of the two climatic elements.

The annual schedule of temperature and precipitation carries in itself rather quantitative than qualitative differences in the three capitals. They can be traced back to astronomical conditions, to the location occupied in the general system of air circulation as well as to geographical conditions influencing the content of vapour.

The annual mean temperature is the highest in Bucharest, whereas it is almost identical in Sofia and Budapest (Figure 1). The difference is more significant in the warm half of the year, whereas in winter the stronger irradiation related to continental conditions counterbalances the warming effect of the more favourable latitude.

Comparing the conditions of precipitation in the three capitals (Figure 2) it is conspicuous that the autumn secondary maximum is present only in Budapest, but it is not manifest in the data of the Romanian and Bulgarian capitals. The difference is not surprising, as frequent Mediterranean cyclones also contribute to the phenomenon in our country, which usually do not reach the two other countries.

Next we turn to the presentation of the territorial distribution of the main climatic elements in the two countries.

Bulgaria is located in the continental climatic zone of long and hot summers. Two thirds of the country is mountainous. Due to the highly structured terrain the climate of Bulgaria is rather variegated.

The annual quantity of precipitation (Figure 3a) is between 700 and 1200 mm in the higher mountainous region (the Balkan Mountains, Rila, Pirin and Rodope), whereas it is between 400 and 700 mm in lower mountains and basins, and it is only between 450 and 500 mm in Dobruja. A large part of the country is dominated by maximum precipitation in early summer and minimum precipitation in winter, both being characteristic features of the continental climate. The late autumn secondary maximum precipitation can be observed along the seaside. To the southwest, in the valley of the Struma River the Mediterranean precipitation is characteristic. Figure 3a shows the speed of the vertical upward current, also influencing precipitation. Its territorial distribution resembles the image of the topography, which is also manifest in the annual distribution of precipitation. At the same time, air arriving from various directions in the region of mountains, where it is forced to move upward, would be arranged in cells moving upward and downward in the annual average.

In Bulgaria the mean temperature of the coldest month is below –6 °C in the mountainous area, it is usually between 0 and –2 °C in the region of mid-high mountains, whereas it is above the freezing-point, around 1–2 °C at the seacoast. In the south-western part of the country winters are as mild as along the seacoast in the valley of the Sturma, surrounded by high mountains. The absolute minimum temperatures in winter are between –25 and –35 °C all over the country, but rarely, one may expect frost of –20–25 °C even at the seacoast. The mean temperature of the warmest month has been generally between 20–24 °C, but it could even reach 25 °C in the valleys of the Sturma and Maritsa. In summer the absolute maximum temperatures may reach 40–43 °C. The water temperature of the Black Sea near the coast would be 22–23 °C in the summer months, and it is ideal for bathing.

Our eastern neighbour, Romania is located at the eastern wing of the Carpathian Mountains. The ranges of the Eastern and Southern Carpathians divide the country not only geographically, but also climatically. The territory inside the ring of the Carpathians, the Transylvanian Basin, together with the regions of the Partium and the Banat mostly belongs to the wet continental zone with a long warm season. The climate in Transylvania is highly variegated due to its articulated topography. The mean temperature of the warmest and coldest months varies depending on the altitudes. The absolute maximum temperature reaches 37–40 °C in areas below 500 m, whereas the absolute minimum temperature of the closed basins can even be as low as –35 °C. Variety is present in precipitation patterns, too. While the annual quantity of precipitation is around 500–600 mm in a large part of the Transylvanian Basin and in the Csík and Háromszék Basins, it may be as much as 800–1400 mm in the Bihar Forest and along the slopes of the Carpathian Mountain range. The variation of precipitation is characterised by a vigorous maximum in early summer and by a winter minimum.

The climate of the Trans-Carpathian regions (Oltenia, Muntenia, Moldva) is typically continental. The short and cold winter is followed by a warm but not too long summer. The absolute annual fluctuation of temperature is significant. In summer temperature may rise to 40–44 °C, whereas it is not rare to have –30 or even –35 °C in winter. The annual average quantity of precipitation is between 400 and 600 mm. The wettest period is early summer, and the least precipitation is to be observed in winter. A characteristic feature of this region is the transitory aridity developing in late summer or early autumn.

In the higher regions of the Carpathian Mountains it is the mountainous climate that clearly dominates. Basically two factors, namely altitude above the sea level and exposure define the climate in this region.

As it can be judged by the colouring of the maps of Figure 4, the presence of the Carpathian range and location related to it is much more important than the north–south differences. This is true despite the fact that the country is located in a strip that is almost twice as broad by latitude as our country or Bulgaria.

Figure 4b presents the calculated values of the horizontal wind speed. The territorial distribution of this value also recalls the picture of the topography, even if the zones of the maximum wind speed follow it with a slight shift. The relationship is simple and obvious: topography as a mechanical obstacle forces the air to circulate upwards, and next, in the new height at a faster speed. (It should be noted that these kinds of speed are not to be measured near the surface, or above forests, for such observations cannot be obtained in the required density, and there is no possibility for the representative modelling of layers a few metres above the many kinds of surface types. Values valid somewhere above the level of the top of trees may be regarded as more or less the same as the characteristic level ten metres above flat terrain, already independent of landmarks.)

Regional Tendencies and Forecasts

Before discussing regional changes, a brief illustration of the possible continuation of global warming is given (Figure 5), which also shows how reliably we can delineate the specificities of its local appearance and their extent.

Figure 5a demonstrates that the global climate models can assess the scope of change in global average temperature caused by the alteration given in time of the – presumably – known external factors governing the climate. The Figure verifies for the past one and a half centuries that the observed changes and those simulated by the model have been essentially parallel to each other.

Based on Figure 5b we may rest assured, that even if the ocean conveyor belt stopped at a certain point of warming (for the reasons see: Czelnai, R., 1999 among others): its consequence would not be an Ice Age, but a distribution of temperature largely different from the present one, in which the present image of the cyclone tracks would greatly change, but the sign of temperature changes would be positive everywhere. For the increasing greenhouse effect should also be taken into consideration in the process that would presumably make the conveyor belt stop.

The next two elements of Figure 5 prove that these models are not yet able to reliably assess the regional details of climate change. Figure 5c shows that even the reconstruction of the present zonal distribution of atmospheric pressure on sea level is done with great differences by the simulation of the models among themselves as well as in comparison with the observed distribution.

Figure 5d compares changes obtained in nine climate models. Two sets of 36–36 correlation coefficients can be obtained by pair-wise comparison of the changes of temperature and precipitation in corresponding grid points. The correlation of the fields of temperature changes between 0.2 and 0.8 is more or less acceptable, but the values of changes in precipitation below 0.4 (and even negative in two comparisons!) clearly indicate that the global climate models are not yet (in 2001) suitable in this respect to offer an ultimate specification about the regional features of change, among others to be able to develop the required adaptation strategies.

Despite the criticism spelt out we present the changes of temperature and precipitation by 2020 on the basis of the model calculations in keeping with the SRES A2 emission scenario (IPCC, 2001). This scenario postulates a heterogeneous economic development and demographic change, as a result of which the average temperature models of the Earth would exceed the values of 1961–1990 by 0.6 °C in 2020.

As shown by Figure 6, temperature would grow everywhere in Europe, and that too, generally more steeply than the average of the Earth. As the three models are arbitrarily taken out of the available results, and because their similarity even according to Figure 5d is only partial, it is not worth preparing a more comprehensive, numerical assessment for the two countries.

Precipitation would clearly be reduced in four maps out of six as shown by Figure 7, whereas change is of opposite direction indicated by the two winter figures in the two countries. Accordingly, a more limited access to water should be expected in the region by 2020, which would be particularly adverse given the growing temperature.

The question is whether the expected warming and decreasing precipitation estimated in calculations can be demonstrated by the observations of the past decades, when the Earth as a whole had been unambiguously warming (Figure 5a).

We answer to this question on the basis of the third assessment report of the IPCC (2001) indicated in Figure 8, as well as on the basis of our own calculations (Mika and Bálint, 2000) given in Figure 9.

It is clear in Figure 8 that the annual mean temperature had grown everywhere in Europe during the period between 1976 and 2000. This estimate shows almost redoubled warming (with minor differences it is 0.6–1.0 °C) in the territory of Romania and Bulgaria, but it has not considered every station of the two countries and has not applied homogenisation for the series of data, compared to the average of the Earth.

The sign of changes in precipitation is by far not so clear on the same Figure. One of the reasons is that the spatial resolution of the processing is rather limited and the number of precipitation-measuring stations is also far lower than one could obtain on the basis of all the national data bases united. The changes of precipitation having an opposite sign (showing a coherent spatial order) could just as well be real ones, for the changes of atmospheric circulation parallel to warming are not everywhere of the same sign. (We should consider that atmospheric pressure cannot change everywhere in the same direction, as the total mass of the atmosphere is constant.)

According to the right side of Figure 8 the sign of precipitation change would be positive in a larger proportion of the two countries. This figure indicates it particularly for the northern part of Romania, based on precipitation data observed between 1976 and 2000. The two grid points falling on the territory of Bulgaria show a change of opposite sign.

As far as Romania is concerned, we have reached an opposite result in our study on the basis of 76 precipitation time series from the 25 years between 1974 and 1998. In that study (Mika and Bálint, 2000) we estimated the b regression coefficient of the Y=a+bX linear relationship with the so-called method of instrumental variables (Körösi et al. 1990). In our procedure we regarded time as the instrumental variable in the given period of linear global temperature trend, which satisfies the conditions of the selection of instrumental variables. These criteria are: a) significant correlation with the values of the independent variable; b) no-correlation with the faults of the independent variable; c) no-correlation with the residual values of the dependent variable. With the help of Z instrumental variable the regression can be estimated as the quotient of co-variances: b=cov(Y,Z)/cov(X/Z). In the 25 years studied the linear trend of the hemispheric temperature is 0.026 K/year, the correlation is 0.825, in other words, condition a) is definitely met. (Condition c) derives from the strong correlation between hemispheric temperature and time, whereas we have to accept condition b) on the basis of professional outlook.)

The regression coefficients gained by this method in Figure 9 are given as percentage change related to 0.5 °C warming. Accordingly, the territory of our country, with the exception of the Small Plain and the Northern Central Mountains, is characterised by a few per-cent fall in the amount of precipitation. East of us the coefficients are negative, whereas they are positive to the west. The change in the winter half-year is clearly negative in Hungary, with characteristic values between –10 and –20 per cent. East of Hungary the reduction of precipitation would be vigorous, reaching even –30 per cent in some places. Growth may be found west of Hungary, in the Alps.

All in all an unambiguous decrease of precipitation in the Transylvanian regions relates to the present subject matter, in other words, it is a result contrary to the content of Figure 8, based on a rare network of stations.

In summary both the trends of temperature as well as precipitation trends accepted for a more comprehensive data base confirm the conditional forecasts derived from the models for the expected climate of Romania and Bulgaria in the current warming period.

Geographic Analogy: Links to Some Hungarian Scenarios

The present climate of several areas in the two countries corresponds in its broad outlines to the image one may assume for the territory of Hungary on the basis of certain statistical relationships and large-space model estimates in some future phases of global warming.

Not giving the details of the methodology of calculations for Hungary, and the results of scenarios, we present the expected changes of temperature and rainfall on the basis of our earlier papers (Mika, 1993, 1996, 2001) in Table 1, and also illustrate them on a map in Figure 10. The bottom line of the Table lists territories that are analogous with the climate, where today’s climate is like the one expected to prevail in Hungary as a result of 0.5–4 °C of global warming.

Even though geographic analogies produce tangible opportunities for linking the Balkan region and our country, moreover, even for the future adaptation of experiences in cultivation, way of life, etc., this possibility should be handled with caution for two reasons. One is that the methodology of the Hungarian climatic scenarios is not fully developed, primarily because of the limitations of direct physical modelling. Practical procedures based on the above rough-dissolution climatic models and on projecting into the future diagnostic relationships valid in the past, even though they yield consonant results, can be regarded as first approximations for lack of adequate explanation of dynamics. The other reason of the necessary caution is that there can be differences in many details between the present climate of the Balkans and the future climate of Hungary even if two conditions of the analogy are met, namely the temperature of extreme seasons and the realisation of the annual quantity of rainfall. Obviously nothing guarantees that there is a climate presently anywhere on Earth that could be expected in the various regions of the future Hungary.

In summary the geographic position of the two countries under survey differs only slightly from that of Hungary. As a result the annual course of the climate of the three capital cities located on flatland is similar. But the mountain ranges of the two other countries divide the territorial distribution of temperature and precipitation into segments. In addition to indicating it we have also devoted particular attention to how the climate of the two countries may change if global warming continues. Despite a broad belt of uncertainty it is probable that the amount of summer precipitation of the region would fall while warming would be faster than the average of the Earth, but the sign of this change in winter is ambiguous. Linking the present climate of the two countries to Hungarian scenarios we have risked the methodological possibility of looking for analogous regions from the present climate of the two countries for our future climate to the environmental and economic impact studies.

References

Bartholy J.–Mika J.: Időjárás és éghajlat – cseppben a tenger? (Weather and Climate – a Sea in a Drop?) Magyar Tudomány, 2005. No. 7. 789–796.

Czelnai R.: Világóceán. Modern fizikai oceanográfia. (World Ocean. Modern Physical Oceanography). Vince Kiadó, Budapest, 1999. p. 182.

IPCC, 2001: Climate Change 2001: The Scientific Basis. (Eds. Houghton J. T. et al.) Cambridge University Press, Cambridge, UK and New York, N. Y. USA, 881 p. http//:www.ipcc.ch

Körösi G.–Mátyás L.–Székely I.: Gyakorlati ökonometria (Practical Econometry). KJK, Budapest, 1990. p. 481.

Mika J.: Az Alföld éghajlatának megváltozása a globális klímaváltozással összefüggésben. (Changes of the Climate of the Great Plain in Relation to Global Climate Change). Alföldi Tanulmányok XV. 1993. 11–30.

Mika J.: Éghajlati forgatókönyvek. In: Változások a légkörben és az éghajlatban. (Climatic Scenarios. In: Changes in the Atmosphere and in the Climate). (Ed. Mika, J.) Természet Világa Különszám, 1996. 69–74.

Mika J.–Bálint G.: Rainfall scenarios for the Upper-Danube catchment. Proc. XXth Conf. Danubian Countries, Bratislava, Slovakia, 4–8 Sept., 2000. CD-ROM, 990–995.

Mika J.: Új kutatások a globális klímaváltozás, valamint ezek regionális sajátosságai és hatásai terén. In: Berényi Dénes jubileumi ünnepsége előadásai (New Researches in the Field of Global Climate Changes, and its Regional Specificities and Effects. In: Presentations at the Jubilee Celebrations of Dénes Berényi). Debrecen, 2001. 185–202.

Mika J.: Globális klímamodellek. Klímaváltozás – hazai hatások. (Global Climate Models. Climatic Change – Domestic Effects.) Természet Világa Különszám. 2004. 33–36.

Mika J.: Globális éghajlatváltozás – két IPCC jelentés között. In: Éghajlatváltozás a világban és Magyarországon. (Global Climate Change – between Two IPCC Reports. In: Climate Change in the World and in Hungary). (Ed. Takács Sánta). Alinea Kiadó, Budapest, 2005. 83–103.

Wood, R. A.–Vellinga, M.–Thorpe, R.: Global Warming and THC stability. Phil. Trans Roy. Soc. A, 361, 2003. 1961–1976.