*3.5. The Temporal Dynamics of Drought*

Drought was practically absent in the first four reporting periods, i.e., 21 March to 20 May up to 21 April to 20 June. CWB values in these periods were several tens of mm with the highest value of −48 mm in the first period, and the area at risk of drought ranged from a fraction of a percent to 2.6% in the fourth reporting period. From the fifth reporting period, i.e., 1 May to 30 June, the mean CWB value for Poland decreased from −87 mm, through subsequent periods −94 mm and −133 mm to reach the highest value of −173 mm in the eighth reporting period from June to July. In the following reporting periods, CWB values systematically decreased from −134 mm, −106 mm, −91 mm, −54 mm, and −44 mm to reach −20 mm in the last reporting period. The percentage of area at risk of drought increased at a similar rate from a fraction of percent in the first three reporting periods through several to a dozen or so percent in periods 4–6, i.e., 21 April–20 June to 11 May–10 July, to reach 28% in period 7 and the highest value of 42% in period 8. In periods 9 and 10, i.e., 11 June to 10 August and 21 June to 20 August, it was already 18% and 6%, respectively, in the following periods a few percent and fractions of a percent. The highest CWB values occurred in the Lubuskie Lake District and systematically increased from −120 mm in the first reporting period to −240 mm to −279 mm in the eighth period of the highest drought intensity. In subsequent periods, they decreased systematically to reach the value from 100 mm to −159 mm in the last reporting period. The percentage of the area endangered by drought in the three lubuskie, łódzkie, and wielkopolskie voivodeships (most endangered by drought) increased from a fraction of a percent to several percent in the first three reporting periods to reach the values of 52%, 73%, 97%, and 56% in periods 6–9, respectively (Figure 7).

**Figure 7.** Temporal dynamics of drought in 14 IUNG reporting periods based on climatic water balance (CWB) values and percentage of area at risk of drought average values for Poland without regional differentiation.

Explanations: reporting period 1 (from 21 March to 20 May), reporting period 2 (from 1 April to 31 May), reporting period 3 (from 11 April to 10 June), reporting period 4 (from 21 April to 20 June), reporting period 5 (from 1 May to 30 June), reporting period 6 (from 11 May to 10 July), reporting period 7 (from 21 May to 20 July), reporting period 8 (from 1 June to 31 July), reporting period 9 (from 11 June to 10 August), reporting period 10 (from 21 June to 20 August), reporting period 11 (from 1 July to 31 August), reporting period 12 (from 11 July to 10 September), reporting period 13 (from 21 July to 20 September), and reporting period 14 (from 1 August to 30 September).

### *3.6. Ranking of Voivodeships by Area at Risk of Drought*

Despite slight differences in the order of voivodeships for average data from all reporting periods and from the eighth period with the highest drought intensity, a clear regularity was observed. It was possible to distinguish four groups of voivodeships with the highest, medium, and lowest drought risk, and one voivodeship in which drought practically did not occur. Voivodeships with the highest drought risk included the lubuskie, łódzkie, and wielkopolskie voivodeships. The voivodeships with a medium drought threat were the dolno´sl ˛askie, kujawsko-pomorskie, lubelskie, mazowieckie, opolskie, ´sl ˛askie swi ˛ ´ etokrzyskie, and zachodnio-pomorskie voivodeships. The voivodeships with a low drought risk included małopolskie, podkarpackie, podlaskie, and pomorskie. In the warmi ´nsko-mazurskie voivodeship, drought practically did not occur (Table 2).

**Table 2.** Ranking of provinces in terms of area at risk of drought (%) at the area of the voivodeship at draught risk.


Source: Agricultural drought monitoring system in Poland IUNG (2019).

#### *3.7. Sensitivity of Crops to Drought*

Analysis of areas covered by drought, both in the whole of Poland and in the three voivodeships most threatened by drought, i.e., the lubuskie, wielkopolskie, and łódzkie voivodeships, revealed clear differences in the drought sensitivity of individual crops. The crops most threatened by drought are fruit bushes, spring cereals, maize, and legumes. For Poland, the percentage of these crops threatened by drought ranges from 16–18%, while, on the area of the three voivodeships most exposed to drought, it is as high as 42–49%. Tobacco, winter cereals, field vegetables, and strawberries are among the crops which are moderately sensitive to drought. On the other hand, the crops least exposed to drought are rape and colza seed, hops, potatoes, fruit trees, and sugar beet, and the percentage of the area of these crops affected by drought is 2–8%, respectively, on the area of Poland as a whole, and 11–26% on the area of the three voivodeships mentioned. In the eighth driest reporting period, from 1 June to 31 July, the majority of crops in the lubuskie, wielkopolskie, and łódzkie voivodeships were almost completely threatened by drought (Table 3).

**Table 3.** Sensitivity of crops to drought. Percentage of drought incidents in 14 consecutive reporting periods and during the period of maximum drought intensity.


Source: agricultural drought monitoring system in Poland IUNG (2019).

#### *3.8. Hydrological Aspects of Drought*

According to the classification of Kaczorowska [4], which assesses the deficiency or excess of precipitation in relation to the perennial norm (1971–2000), the year 2019 was classified as normal [Bulletin of the National Hydrological and Meteorological Service No. 13/215 2019]. Annual precipitation on a national scale, determined by measurements from 52 synoptic stations, amounted to 556 mm, corresponding to 91.7% of the multi-year value (1971–2000). In the central part of Poland, 2019 was classified as dry, locally even as very dry, and in the rest of the country as normal, only locally in the north and south as wet. The hydrological year 2019 with a total outflow of Polish rivers equal to 41.9 km<sup>3</sup> (with an average in the multi-year period 1951–2018 equal to 60.4 km3) was classified as a dry year [Bulletin of the National Hydrological and Meteorological Service No. 13/215 2019]. The outflow in 2019 was, therefore, about 2/3 of the multi-year average outflow. In the distribution sequence from the period 1951–2019, which contains 69 years, where the years are arranged in the order of increasing total annual outflow of Polish rivers, the year 2019 occupies the 5th place after the driest the years, i.e., 1954, 2015, 2016, and 1952. In the first and second hydrological half-years, lower values of outflow occurred in the Odra basin than in the Vistula basin. The outflow of the Pomeranian rivers was, like other rivers, lower than the norm, and was generally relatively higher than the outflow of the rivers of the Vistula and Oder basins. Outflow of the Pomeranian rivers, like other rivers, was lower than the regular values; however, in general, it was relatively higher than the outflow of the rivers of the Vistula and Odra basins [Bulletin of the National Hydrological and Meteorological Service No. 13/215 2019] (Table 4).

**Table 4.** Months with the lowest flow in 2019 compared to the 1951–2018 multi-year period.


\* Flow values are affected by water management in the reservoir. Source: Bulletin of the National Hydrological and Meteorological Service No. 13/176 2016, 13/189 2017, 13/202 2018, 13/215 2019

After analyzing 20 river flows of very different river basins [Bulletin of the National Hydrological and Meteorological Service No. 13/215 2019], in almost half of them, i.e., in nine, the drought accelerated the date (month) of the lowest flow. Thus, this acceleration was observed in three cases by one month (two cases from September to August and one case from August to July). In four cases, it was an acceleration by two months (in two cases from October to August and in two cases from September to July). In two cases, the acceleration of the lowest flow was by three months, i.e., in all cases from October to July. These flows constituted on average of 41% in average perennial flows, a maximum of 52% (Vistula–Warszawa), and a minimum of 36% (Odra–Nowa Sól). It is noteworthy that all cases of minimum flow acceleration only occurred in central-western and south-western Poland with the highest air temperature. In seven cases, apart from those analyzed above, minimum flows occurred in the same months as the multi-year averages, but their values were, on average, smaller by half. In the case of the Skorogoszcz gauge on the Nysa Kłodzka River, flow values were affected by water management in the existing reservoir. Indeed, at the gauge mentioned, minimal flows did not occur in July and August, like most of the analyzed gauges, but in November. Low flows also occurred in October, December, and August. Water management in the reservoir, in this case, disturbed the natural rhythm of flows.

#### *3.9. Hydrogeological Aspects of Drought*

In the hydrological year of 2019, the level of the groundwater table was lower than the monthly average for many years (1991–2015). In waters with a free well, the largest deviation was recorded in August (about 31 cm below the average for many years). In waters with a tight well, the highest values below the average were recorded in the period of August–October where it was about 32–34 cm below the multi-year average. In July, the state hydrogeological service declared a state of hydrogeological emergency due to the phenomenon of a very intense hydrogeological low. In August, an extension of the area of the phenomenon was found, and, by the end of the hydrological year, there were no grounds for revoking the state of hydrogeological emergency [Hydrogeological Annual Report Polish Hydrogeological Survey].

In the lowland belt, where the drought was the most intensive, the mean monthly groundwater levels of both free and tight groundwater tables were lower than the monthly averages for the whole hydrologic year. The highest deviation of free-bore waters was recorded in August (36 cm below the average), and, by the end of the year, they were over 30 cm below the average for individual months of the studied period. Similarly, it was the case for waters with a tight well, the average monthly groundwater level was at a level lower than the monthly averages (this difference was 9 cm below the average in January), and then increased from month to month to 45 cm below the average in October, representing the largest deviations of waters with a tight well [Hydrogeological Annual Report Polish Hydrogeological Survey].

#### **4. Discussion**

A direct comparison of the IUNG method (CWB in 6-decade periods, step every decade) with the method used in the study (CWB in monthly periods), i.e., concerning the months of the most intensive drought in June, July, and August, is not possible due to different comparison periods. The first two decades of June in the IUNG method fall within the 6th reporting period, ie 11 May to 10 July. The second and third decades of June fall within the seventh reporting period, covering the period from 21 May to 21 July. Therefore, the average of the reporting periods for June (from reporting periods 6, 7, and 8), for July (from reporting periods 9, 10, and 11), and for August (from reporting periods 12, 13, and 14) were conventionally taken for the estimation of CWB values.

The results vary slightly, though there was the least variation in June. June was the warmest month since at least 1951 with a temperature deviation from the national norm of 5.0–6.0 ◦C and the driest relatively month of the year in 2019, and was very dry on the verge of extremely dry which was the reason for the very low CWB values [Bulletin of the National Hydrological and Meteorological Service No. 6/208 2019]. Smaller differences occurred in August and the largest in July. The differences may be caused, on the one hand, by the different way of estimating values for individual months as an average of the three reporting periods and, on the other hand, by the extension of large negative CWB values from June to the July (post-June) using the IUNG method to determine CWB for 6-week periods (Table 5).

**Table 5.** Comparison of climatic water balance (CWB) values (mm) estimated from IUNG studies and those obtained in this study.


The drought of 2019 was preceded by the drought of 2018. The analysis of consecutive communiqués of the IUNG agricultural drought monitoring system from 2020 shows that, in that year, drought was indicated, especially in north-western Poland voivodeships; in the wielkopolskie voivodeship during all 14 reporting periods; and in the lubuskie, pomorskie, and zachodniopomorskie voivodeships during 13 periods (except the last reporting period). In 2021, in the fifth reporting period from 1 May to 30 June, drought of various intensity was observed in all Polish voivodeships. The analysis of the causes, the course, and the consequences of the dry years of 2018–2021 requires a separate study.

The drought of 2019 has marked itself with varying intensity across Europe. The manifestations and impacts of drought are reported from countries, such as Belgium, Czechia, Finland, France, Germany, Greenland, Italy, Luxembourg, Netherlands, Norway, Poland, Spain, Sweden, Switzerland, and the United Kingdom [64]. A historical reconstruction of the 254-year climate database [65,66] indicates that many years with similar precipitation anomalies occurred in the summer months, but 2018–2019 saw two of the three warmest summer periods in the cited period. The third year when average summer temperature anomalies over Central Europe reached record extreme conditions of over 2.0 ◦C was 2003. Of the three cited years with the highest air temperature anomaly in Europe, in 2019, the strongest impact of the thermal anomaly was marked in the Central European region including Poland where the increase in temperature was accompanied by a simultaneous significant reduction in summer precipitation, leading to extreme drought conditions.

The comparison of the values of climatic water balance in the months of June, July, and August in the analyzed three extreme years in terms of temperature in the area of Poland shows that, in 2003, these values were −100, −21, and −79 mm, respectively. In 2018, the values averaged for Poland were −77, −36, and −80 mm, respectively. In 2019, the year under analysis, these values were, −129, −64, and −53 mm, respectively, so the highest were in June and July, and were slightly lower only in August. If the values of climatic water balance for the analyzed three months were averaged, then in the next extreme years, in terms of temperature, the values would be −67 mm in 2003, −64 mm in 2018, and −82 mm in 2019. Thus, it can be concluded that, in the light of the values of climatic water balance in Poland representing a significant area of central Europe, this was the most intense drought in the reconstruction of the last more than 250 years of climatic data.

Analysis of drought frequency over a multi-year period shows that the frequency has increased since 1950 throughout southern Europe and most parts of central Europe, while it has decreased in many parts of northern Europe [67,68]. Other drought indices, including drought severity indices, also show significant increases in the Mediterranean region and parts of central and south-eastern Europe, and decreases in northern Europe and parts of eastern Europe [69–72]. Projections for the period 2041–2070 compared to 1981–2010 for two emission scenarios, i.e., RCP4.5 and RCP8.5, indicate an increase in meteorological droughts in most of Europe, especially in southern Europe, while decreases in droughts are projected only for a limited part of northern Europe. The changes are most pronounced for the high emissions scenario (RCP8.5) and somewhat smaller for the moderate scenario (RCP4.5) [73].

#### **5. Conclusions**

The summer drought of 2019 was not only caused by a shortage of precipitation, but above all by extremely high temperatures, especially in June with a record deviation of 5.0–6.0 ◦C and in August when deviations exceeded 2.0 ◦C. Anticyclonic situations prevailed especially in June (59%) and August. In June, the advection from south direction was 2.5 times higher than the average of 8%.

In the light of days with low precipitation frequency (frequency lower than average), 41% of the duration of summer drought lasted from 29 May to the first days of September. The longest, i.e., 90 days and more, occurred at stations located in the Central Poland Lowlands (up to Warsaw) and the southern part of the South Baltic Lake District. On the outskirts of this driest area, the number of such days with a below-average precipitation frequency was systematically lower. Such a period with lower precipitation frequency basically did not occur only in the area of the Gulf of Gda ´nsk, the Eastern Baltic Coast and the Lake Districts.

In the summer season of June–August, the lowest precipitation amounts, constituting 30–60% of the norm were recorded in the South Baltic lakes belt and Central Polish Lowlands with the minimum of 30.1% in Pozna ´n. The highest values of precipitation were noted in the eastern part of the South Baltic Coast, the Eastern Baltic Coast, and the Podlasie–Byelarus Uplands with 100–125% of the norm. The average precipitation in Poland in summer was only 66.5% of the norm.

The values of climatic water balance (CWB) calculated by the Institute of Soil Science and Plant Cultivation (IUNG) method for particular months of June–August for the area of Poland amounted successively to −129 mm, −64 mm, and −53 mm with minima, respectively, of <sup>−</sup>176.6 mm in Pozna ´n, <sup>−</sup>109.8 mm in Swinouj´ ´ scie, and from <sup>−</sup>92.4 mm to −100.6 mm near Słubice, Gorzów Wielkopolski, and Pozna ´n.

Summer drought in the light of the area of Poland threatened by drought increased gradually exceeding the values of several percent of the area of Poland in the periods from the beginning of June to 10 July, with the maximum at the end of June—42% of the area of Poland covered by drought. Most threatened by drought were the wielkopolskie, lubuskie, and łódzkie voivodeships, in which the percentage of the area threatened by drought during its highest intensity in the eighth reporting period (1 June to 31 July) amounted to 92%, 88%, and 81%, respectively.

The sensitivity of individual crops to drought is evident. The least resistant to drought are: fruit bushes, spring cereals, maize for silage and grain, and legumes. Medium drought tolerant are: winter cereals, tobacco, field vegetables, and strawberries. The most resistant are hops, potatoes, rapeseed and canola, fruit trees, and sugar beet.

The summer drought accelerated the timing of flow minima, which typically fall in the autumn months, by 1 to 3 months or halved the minimum flows without changing the timing of their occurrence. In the hydrological year 2019, the level of the groundwater table was lower than the average monthly level in many years. In waters with a free well, the largest deviations were recorded in August, and, in waters with a tight well, the largest values below the average were recorded in the August–October period.

Although contemporary trends and scenarios of precipitation changes at the end of the 21st century for Poland do not predict a decrease in precipitation and even a small increase of about 5–10%, it cannot be said that we will not be threatened by droughts. These will be droughts caused not so much by a lack of precipitation but by high air temperature combined with low precipitation (the increase of which in Poland has been documented). The phenomenon of drought will also be aggravated by the observed increase in the coefficient of variation of precipitation in Poland.

**Funding:** The research was partially financed by Ministry of Science and Higher Education in Poland.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.
