FORMULATION OF THE PROBLEM

The COVID-19 pandemic has completely changed the entire world and our understanding of its sustainability. It divided the current stage of world development into the pre-COVID and COVID eras. It was a very significant impact on the tertiary sector of the economy, especially tourism, the hotel business, and the entertainment industry, as well as on communications, including transport connectivity and people’s mobility. First of all, people themselves suffered, and secondly, the economy.

It is still difficult to estimate the scale and nature of the impact of the COVID-19 pandemic on the territorial structure of the economy as a whole, its individual industries, people’s lives, the settlement pattern, and spatial structure of urban agglomerations, in particular, due to the lack of the necessary territorially fractional statistical data. From the viewpoint of a transport geographer, it is important to understand how strong the impact of the pandemic on transport was, i.e., how significant was the decline in the volume of traffic by each type of transport in different countries and macroregions of the world and in certain areas within countries, and how was this reflected in the territorial structure of transport. Unfortunately, such an analysis is not yet possible due to the extremely incomplete and fragmentary statistical information on the transport system in a territorial context.

Air transport in this respect yielded better results than did other modes of transport, since at most airports, by fall 2021, accurate information on their passengerFootnote 1 and cargo turnoverFootnote 2 was published for the 2020 in its entirety versus 2019 (and for a number of airports even for every month of 2019 and 2020). This allowed the author to carry out this study.

The pandemic led to restrictions of people’s movement both within and between countries, as a result of which the passenger traffic by air transport sharply decreased, the level of aviation mobility of people decreasedFootnote 3, and the degree of air connectivity between cities and countries decreased. All these changes had pronounced spatial differences.

Revealing the geographical features of the impact of the pandemic since its beginning has become an important topic of Russian economic and geographical research. By summer 2020, a series of general articles appeared on the impact of COVID-19 on the structure of the economy and society as a whole. A special section of the journal Sotsial’no-Ekonomicheskaya Geografiya. Vestnik Assotsiatsii Rossiiskikh Geografov-Obshchestvovedov, no. 1 (9), 2020 contains very short essay articles on this topic (Gerasimenko T. and Gerasimenko A.; Druzhinin; Zyryanov; Kagansky; Kolosov; Kuznetsova; Rodoman; Shuper and others), which have a preliminary evaluative character of this new phenomenon. The most interesting among them is a publication by V.L. Kagansky, in which he analyzes the possible spatial and functional consequences of the impact. A number of publications that appeared several months later analyzed the diffusion of the pandemic, the factors of its spread, and regional impact on the Russian economy (Zemtsov and Baburin, 2020a, 2020b; Zubarevich and Safronov, 2020).

In 2021, new articles on this topic were published in Russian geographical journals. In (Panin et al., 2021), a cartographic analysis of the spatial patterns of the spread of the COVID-19 pandemic in Russia is carried out; it states that the three initial centers of its distribution were the Moscow region, the oil and gas producing north of Western Siberia, and the North Caucasus. The main factors of the rapid spread of COVID-19, from the viewpoint of the authors, were not only transport and logistics parameters, but also a high proportion of the creative class in the Moscow region; shiftwork flows and overcrowding of shift settlements in the Yamal-Nenets Autonomous Okrug; and increased contacts and weak healthcare system in regions of the North Caucasus. Makhrova and Nefedova (2021) consider the possibilities of a transition from seasonal dacha mobility to real suburbanization and deurbanization in areas with different degrees of remoteness from Moscow under new conditions of quarantine restrictions.

Abroad, studies have also appeared on analysis of the geographical factors involved in the spread of COVID-19. A special issue of the Dutch journal Tijdschrift voor economische en sociale geografie no. 3, 2020, published a series of articles on the relevant topic (Geography …, 2020). Kuebart and Stabler (2020) use a spatial diffusion model to study the spread of COVID-19 in Germany.

Chen et al. (2021) apply a gravity model to study the spatial diffusion of COVID-19, which spread from Wuhan to cities in Hubei province in China. The simulation results showed that the size of provincial cities and distance from them to Wuhan influenced the total number of confirmed cases of the disease and that Wuhan was the main source. Thus, the spread of the epidemic was hierarchical, while the immediate neighborhood of cities with each other did not much matter.

Sigler et al. (2021), using regression analysis, obtained the following conclusions: the values of human development indicators and total population predict well the spread of COVID-19 in countries with a large number of reported cases (per 1 mln inhabitants); larger household sizes, older populations and more intense human interactions predict the spread of COVID-19 in countries with low reported cases (per mln inhabitants). Population density and other characteristics of the population, such as total population, proportion of older persons and household size, are relevant explanatory indicators in the early weeks of the epidemic, but have a smaller impact on the rate of spread of COVID-19 over time. In contrast, the influence of interpersonal communication and out-of-store trade increases over time, indicating that higher human mobility may best explain the persistent spread of the disease.

In (Ascani et al., 2020), the spread of COVID-19 in Italy at the regional level (NUTS 3) is analyzed in terms of economic geography. The bright spatial unevenness of the spread of the disease indicates that the coronavirus infection hit the cores of the most economically developed regions of the country the hardest. The specialization of a particular region in a certain type of economic activity that is geographically concentrated turns out to be a means of transmission of the disease. This leads to the formation of center–periphery spreading of COVID-19, where the disease can follow along the main axes of the regional economic landscape.

In (Ramires-Aldana et al., 2021), the spread of COVID-19 across Iran at the ostan (province) level from February 19 to March 18, 2020 is described by a series of cartograms of increase in the number of cases compiled by the authors. The highest morbidity rate was found in the urban agglomeration of Tehran. Urban provinces with older populations and higher average temperatures had significantly more COVID-19 cases. The higher the population’s level of literacy (including health), the fewer their number.

Florida and Mellander (2021) investigate the geographic drivers of COVID-19 in Sweden. These include population density, household size, existence of air transport, income, race and ethnicity, age, political affiliation, temperature and climate, isolation, and physical distance from each other. This study examined the impact of some of these factors on geographic differences in the number of COVID-19 cases and deaths, both by municipalities and provinces (län). It turned out that the geographical variations in COVID-19 are largely associated with variables such as population density, population size, socioeconomic characteristics of localities, and household size. The geographic variability of COVID-19 across Sweden has been found to be large, but highly random.

At the very beginning of 2021, the first, most general articles on the impact of the pandemic on air transport worldwide during 2020 appeared in a special press dedicated to air transport problems (Air Passenger …, 2021; Dunn, 2021, 2020; Worst Year …, 2021), but they lack analysis of the spatial differentiation of this process. There are still very few publications on the geographical analysis of the dynamics of air passenger traffic in the context of the COVID-19 pandemic, since there is still no corresponding territorially fractional statistical information for 2020 in the context of countries and their individual parts. For this reason, this issue has hardly been studied. Suau-Sanchez et al. (2020) assesses the medium- and long-term impact of the pandemic on air transport based on interviews with airline executives. However, this article is not so much geographic as sociological in nature.

This article partially fills the existing gap in the geographical analysis of this phenomenon. The objectives of the article are to identify the process of the spread of the COVID-19 pandemic across Europe and Russian regions from a transport–geographic viewpoint, as well as territorial differences in the dynamics of passenger turnover at airports in European countries and Russian regions in the first year of COVID 2020 versus pre-COVID 2019. It is important to study and understand what these differences are from place to place, where this decline was stronger and weaker, and why in some places the decline was very large, while in others it was medium and small. Geographic concepts (including the theory of spatial diffusion of innovations) make it possible to identify such differences.

MATERIALS AND METHODS

To analyze the dynamics of air traffic, statistical data on passenger turnover for almost 300 large and medium-sized airports in Europe and Russian regions for 2019 and 2020 were used, which were collected from a large number of sources: annual reports of individual airports, groups of airports, national aviation agencies, and departments of individual countries. The first reports (including monthly) for 2020 were published in January 2021, but most appeared on the Internet only in May–July 2021.Footnote 4 Thus, information from different sources was combined and systematized.

Airport passenger turnover proved best indicator of spatial activity and passivity of people during COVID-19: if an airport was closed, it meant complete lockdown in a given city and the surrounding area; if the number of passengers was insignificant compared to the pre-COVID year, then this indicated a partial lockdown; the smaller the decrease in passenger traffic through the airport during the year, the weaker were the restrictions on movement. Based on the dynamics of passenger turnover at most airports, one can judge not only its nature, but also geographic differences.

The values of the total passenger turnover of all airports of each European country for 2019 (the last pre-COVID) and 2020 are compared. The changes are measured absolutely (growth/decline) and relatively, as: (1) the ratio of growth/decline to value in 2019,%; (2) the ratio of values in 2019 to 2020, times. In addition, the time lag of the depth of retrodegression was calculated (how many years ago the passenger turnover in a COVID year was “set aside”).

To break down changes in airport passenger turnover into types, the author used his own methodology developed earlier for quantitative assessment of the degradation of passenger air connectivity at Russian airports in 1990–2006 (Tarkhov, 2015, p. 124), where the types of growth and decline in air passenger traffic were empirically identified. The quantitative parameters of the types are presented in Table 1 (gradations for the third indicator of dynamics are added).

Table 1. Types of airport passenger turnover dynamics in context of radical socioeconomic crises
Full size table

The author has calculated the indicators of the passenger turnover dynamics at airports in almost all European countries, including Russia, for which complete statistical data were found.

RESULTS

Spatial Distribution of COVID-19 in Europe: Transport–Geographic Analysis

Analysis of electronic information sources (including media publications) revealed the centers of the first cases of COVID-19Footnote 5 and made it possible to trace how coronavirus infection spread across Europe. The SARS-CoV-2 coronavirus was introduced in Europe in the last ten days of January 2020 due to the availability of direct flights between Wuhan (China) and several major European airports (Table 2). Regular passenger flights from Wuhan airport at the end of 2019 flew to Paris (Charles de Gaulle airport), London (Heathrow), Rome (Fiumicino), Istanbul, Moscow (Sheremetyevo), and St. Petersburg; there were also some charter flights to other European airports; there were cargo flights to Luxembourg.

Table 2.   Geographic features of spread of SARS-CoV-2 in European countries (listed in chronological order when first cases were recorded)
Full size table

The first case of COVID-19 in Europe was reported on January 24, 2020, in Bordeaux (France). Then, cases were recorded on January 27 in Germany, on January 28 in Finland, on January 30 in Italy, and on January 31 in the UK, Spain, and Sweden, when Chinese tourists rushed to spend the Chinese New Year holidays in European tourist centers, ski resorts, and shopping centers in France and Italy to take advantage of post-New Year sales. On January 31, the first cases were recorded in Tyumen and Chita (Russia), where residents from China arrived from Wuhan. On February 2, the first case arrived in Belgium on a flight from Wuhan. Thus, from January 24 to February 2, coronavirus infection entered European countries with air passengers from Wuhan (either Chinese tourists or Europeans returning from China).

The main peak of infections occurred first in Italy, since this is the European country most visited by Chinese tourists.Footnote 6

The first two cases recorded in Italy were two Chinese tourists in Rome on January 30, who arrived on January 23 from Wuhan at Milan (Malpensa) airport, then flew to Verona airport and traveled by land to Rome via Parma. However, a massive COVID-19 outbreak occurred almost three weeks later, on February 21–22, and the number of infections skyrocketed. The main centers of the epidemic were the Lombardy and Veneto regions. Airports in Northern Italy became the main centers of transmission and cross-spreading of infection, since they had direct flights with many other air hubs not only in Europe, but also in Russia, Africa, and America. At the largest airports in Lombardy, the infection was picked up by tourists returning to their homeland at the end of February from the alpine ski resorts of Northern Italy, becoming coronavirus vectors to other countries (see Table 2). Since the end of February, northern Italy has become the main source of diffusion of infection. From February 25 to March 13, it spread from Milan airports (Malpensa and Linate) and Bergamo to 28 European countries, i.e., the overwhelming majority.

The second spreading center since the end of February was Innsbruck Airport, where COVID-19 very quickly entered from neighboring Lombardy. Tourists on holiday in the ski resorts of Tyrol flew to all corners of Europe. Via Austria’s airports, the infection first entered Norway and Denmark.

Another source of infection was Iran (coronavirus infection was first brought to this country by a businessman from Qom, who visited Wuhan: the first fatal case was recorded on February 19, 2020), from where COVID-19 together with air passengers (most often Iranian students returning to study in European universities) penetrated four countries of Eastern Europe (Belarus, Latvia, Estonia, and Hungary).

The infected arrived in Montenegro by air from New York and Barcelona.

By road transport, the coronavirus penetrated from Switzerland to Liechtenstein, from Germany to Poland, from Hungary to Serbia, from Italy through the Romanian airport, and then by vehicle to Moldova.

The COVID-19 epidemic quickly turned into a pandemic, affecting all countries without exception. At the end of January, these were large countries (France, Germany, Great Britain, Italy, Spain). In February – early March, due to the rapid spread, medium-sized and then small countries were also added. The last to receive the coronavirus were the states of Eastern and Southeastern Europe (Moldova, Bulgaria, Albania, Kosovo, and Montenegro, March 17), as well as unrecognized post-Soviet republics (Pridnestrovian Moldavian Republic, March 21; Luhansk People’s Republic, March 29).Footnote 7

On August 4, 2021, the total cumulative number of all coronavirus cases during 2020 and the first seven months of 2021 in Europe was 52 068 704.Footnote 8 The number of cases in the first 7 months of 2021 in all European countries as a whole was 2.2 times higher than all of 2020.

The coronavirus came to 8 countries from Wuhan, to 28 countries from northern Italy, 4 from Iran, 2 from Austria, and the rest from others. As Table 2 demonstrates, it reached some countries at the same time from several. In 39 out of 49 countries, the first cases arrived by air transport, which became the main vector of the epidemic. In 24 out of 49 countries, the infection first appeared in their capitals, then from there it began to spread by other modes of transport within them; in 25 countries, the coronavirus was first recorded in noncapital cities.

The objectives of the article did not include analysis of in-country differentiation of the number of cases, but from Table 2, it is obvious that the regions and cities within countries most affected by the pandemic differ greatly by type: in some places, these are the largest cities and their agglomerations, while in others, completely different areas and centers, and still in others, remote outskirts. Nevertheless, even the most superficial comparison of countries in this table shows that the pandemic hit major cities the hardest.

Spatial Distribution of the COVID-19 in Russian Regions

Table 3 shows the geographical origins of the first cases of the disease in Russia. On the first day (January 31), the carriers of the disease were Chinese citizens who had come to Russia; these were isolated cases. A month later, Italy became the main source of infection, from where 32 cases arrived by plane (31 were Russians vacationing in ski resorts in Northern Italy or traveling in Italy). This “Italian inflow” took place from February 27 to March 12. On March 13, the first infected Russian tourists arrived from France and Austria; on March 14–15, this was augmented by those who had returned from Spain and Switzerland.

Table 3.   Geographical features of spread of SARS-CoV-2 in Russian regions in first days of 2020 epidemic (listed in chronological order the first cases were recorded)
Full size table

The first peak of the epidemic was on March 2–8 (coronavirus infection was recorded in nine Russian regions); the second peak, on March 16–19 (it already covered 39 regions, including 26 new ones). Then, the epidemic began to expand rapidly in all geographic directions. The first and second peaks were concentrated in European Russia (ER), subsequent peaks everywhere, including the regions of Asian Russia (AR). By March 10, 2020, cases were recorded in nine regions (including 7 ER and 2 AR); March 20, in 40 (including 30 ER and 10 AR); March 25, in 55 (including 44 ER and 11 AR); March 30, in 71 (including 54 ER and 17 AR); April 5, 79 (including 58 ER and 21 AR); April 17, 85 (including 61 in ER and 24 in AR). Thus, the geographical lag of the spread of the epidemic throughout Russia was 51 days (February 27 – April 17Footnote 9).

Due to its transport remoteness, as well as complete or partial overland isolation, the last regions where the disease arrived were the Tyva Republic, the Nenets and Chukotka autonomous okrugs, and the Altai Republic (April 10–17, 2020). These are predominantly more backward socioeconomic regions with a deep-peripheral transport–geographic location.

The first infected people arrived in Russia by air from China and Italy. In March, the main carriers were Russian tourists returning from cities and ski centers in Northern Italy. Then they from Moscow and St. Petersburg, where they originally arrived, moved by plane and train to the regions, spreading the infection, first of all, to all the largest cities of the country (Nizhny Novgorod, Yekaterinburg, Novosibirsk, Krasnodar, Rostov-on-Don, Samara, Kazan, etc.). At the same time, the disease was transmitted by ground transport passengers in the gravitational zones of the airports of these major cities. Infected people later moved mostly by land transport, but to remote and inaccessible regions, exclusively by air (north of Krasnoyarsk Krai, the Sakha Republic, Magadan Oblast, Kamchatka Krai, Chukotka, Nenets and Yamalo-Nenets autonomous okrugs).

Impact of Coronacrisis on Passenger Air Transport

The COVID-19 pandemic led to a sharp decrease in the air passenger traffic, since the borders of many countries were closed and strict controls on the movement of people were introduced, which significantly reduced the need for air travel and people’s air mobility. Many companies stopped flying their planes or went bankrupt. Some airports were closed (e.g., Paris Orly did not operate from the beginning of April to the end of June 2020), and in the largest, some terminals were closed and the remaining flights are concentrated in one to two terminals. In October 2020, 193 regional out of 740 European airports were close to bankruptcy due to a sharp decline in passenger traffic.

The peak of the pandemic in April–May 2020 led to an almost complete cessation of flights, when passenger traffic was reduced to a minimum (Fig. 1). Most scheduled international flights were discontinued, and several countries have even banned domestic air travel in an attempt to contain the spread of COVID-19. This crisis has hit especially hard on international air transportation. It had a much smaller impact on domestic passenger traffic.

Fig. 1.
figure 1

Density of civil aircrafts in space of Europe: (a) April 18, 2019, 11:40 (number of aircrafts in sky, 3100); (b) April 16, 2020, 11:40 (number of aircrafts in sky, 380). Author A.D. Suzansky, May 25, 2020.

Full size image

Decline in Passenger traffic at Individual European Airports

Airports in Europe were hit hardest in the world by the COVID-19 pandemic, with their largest decline in passenger turnover. It decreased particularly sharply at airports with values of more than 1 mln people in 2020 (Table 4).Footnote 10

Table 4.   Decrease in passenger turnover at European airports (excluding Russia) in 2020 due to COVID-19 pandemic (more than 1 mln passengers), thous. people (listed in descending order of passenger traffic in 2020)
Full size table

The strongest decline of passenger turnover occurred at the largest airports of London–Gatwick, Munich, Barcelona, Frankfurt am Main, and London–Heathrow. A strong decline was typical not only of the largest, but also for some airports serving regional traffic (Graz, Nottingham, Leeds, Cardiff, Southampton) and local centers (Exeter, Newquay, Paderborn). In addition, these were small airports in large air hubs where the main airport ”suppressed” them (for example, Girona near Barcelona; London City and London Southend near other London airports, Glasgow–Prestwick near Glasgow airport); a significant number of flights from small airports in this hub were discontinued, while flights to the same cities were operated from the main airport (for example, Reus near Barcelona).

The decline in passenger turnover was medium or moderate at small airports and in areas where types of land transport are either poorly developed or absent altogether, or airports are very remote from the main area of development with a relatively high population density. Airports located on remote islands or very far from economically developed territories have not lost as many passengers (e.g., Voar in the Faroe Islands, Bornholm, airports on Sicily and Sardinia in Italy, etc.).

A number of tourist airports passed through the COVID crisis relatively unscathed (Gran Canaria, Tenerife Norte, Lanzarote, Fuerteventura—all in the Canary Islands), others, conversely, suffered greatly (Tivat in Montenegro, Dubrovnik and Pula in Croatia, Girona–Costa Brava in Spain).

Let us briefly characterize the peculiarities of the decline in passenger turnover at airports for only a few countries.

Germany. A catastrophic decline in passenger turnover occurred at Paderborn Airport (87%); very strong at the airports of Leipzig–Halle (80%), Münster, Weeze and Nuremberg (78%), Cologne–Bonn, Munich and Hanover (77% each), Dresden and Berlin Tegel (76% each). The smallest decline was observed at the airports of Memmingen (specializing in servicing low-cost airlines near Munich; 60%) and Dortmund (55%). At the same time, the low-cost Hahn (71%) and Weeze (78%) airports lost more passengers than Memmingen.

United Kingdom. Despite a significant share of international passengers, the country’s largest airport, London–Heathrow, lost only 73% of passengers. There were significantly fewer passengers in Southampton (83%), London City (83%), Leeds (81%), Nottingham (81%). Aberdeen Airport was least affected by the pandemic (passenger traffic decreased by only 66%). Catastrophic declines occurred in Cardiff (87%), Glasgow–Prestwick (86%) and Exeter (85%).

Spain. The maximum (catastrophic) decline was typical of the airports of Reus (96%) and Girona (91%), and strong for Jerez de la Frontera (81%), Murcia (80%), Almeria (80%), Palma de Mallorca (79%) and Barcelona (76%). The smallest decline (moderate) was observed at the airports of Gran Canaria (61%), Melilla (55%), Tenerife Norte (52%), La Palma (51%).

In Italy the passenger turnover of airports has decreased by 63–80%, and for the exclusively tourist-oriented Rimini (which serves the seaside resorts of Riviera de la Sole) and Treviso (Venice low-cost airport), by more than 80%. A strong decline occurred at the airports of Rome–Fiumicino (77%), Florence (77%), Venice (76%), Pisa (76%), and Milan–Malpensa (76%), i.e., the main ones serving tourists. However, at the airports farthest from the economically developed north, in the west of Sicily and Sardinia (Trapani, Palermo, Alghero), it decreased much less (55-61%), and at the airport on the small island of Lampedusa, quite insignificantly (36%). The decline has affected airports with a narrow tourist specialization, and it was much less for airports with the common functions of serving the surrounding territories.

For all the airports of France, complete information on the dynamics of their passenger turnover is not yet available. The available data for the ten largest of them reveal that the rate of decline varies from 67 to 71% (the latter figure was noted at the airports of Marseille, Paris–Beauvais, Paris–Charles de Gaulle, EuroAirport, and Bordeaux).

In Greece, airports with largest share of international tourists in their passenger turnover were hardest hit by the pandemic (Chania and Zakynthos, 76%; Santorini and Cephalonia, 75%; Mykonos, 73%; Rhodes, 72%). Skiathos Airport had the maximum decline (80%; Northern Sporades islands). Athens and Thessaloniki, the country’s largest airports, lost 66–68% of their passenger traffic. Some airports in the small islands of the Aegean archipelago, far removed from mainland Greece (Mytilene, Naxos, Milos, Lemnos, Chios), also lost fewer passengers than other large tourist airports in this country.

Most airports of Norway are characterized by a relatively insignificant reduction in their passenger turnover (by 35–60%, moderate and medium declines), but for two serving Oslo, there was a much greater decline (68–69%, strong). In the southern, most developed part of the country, the passenger turnover of airports decreased by 61–70%. The farther north from Oslo an airport is located, the lower the decline in its passenger turnover (maximum by only 25–28%). This is because, in the central and northern parts of the country, air transport remains the only means of fast communication, although there is a network of good roads and most cities have seaports, but the distances to the south of the country are significant. An exception is the main airport of Svalbard (decline of 63%).

In Finland, the decline in passenger turnover is higher in the airports of the largest cities in the south (62–83%), while in the north it is much lower (38–60%), as in the northern airports of Norway. The largest reduction in passenger traffic is characteristic of southern Finnish airports that are simultaneously important rail and road junctions in this most populated part of the country. Helsinki Airport declined 77%; Turku, 75%; and northern Rovaniemi, only 59%.

For Swedish airports, the drop was mostly 67–80%, and the only exceptions were the northern airports of Kiruna (59%) and Luleå (64%) (medium decline). The largest (strong) decline was typical for the three airports of Stockholm: Bromma (80%), Skavsta (76%), and Arlanda (75%).

In Iceland, passenger turnover decreased very strongly (81%) only at the main international airport of Keflavik, while at local airports (Reykjavik, Akureyri), this was not as significant.

Poland. Most of all, the decrease in passenger traffic fell on the airports of Szczecin (83%), Poznan (72%), Wroclaw (72%), low-cost Warsaw–Modlin (72%) and Warsaw–Chopin (71%).

In Ukraine, the decline was not as significant (with the exception of both airports of Kiev): four out of six airports had a moderate and weak decline, which was not observed in the rest of Europe.

Decline in Passenger Turnover at Individual Airports in Russia

Unlike European airports, Russian airports had a moderate annual decline, i.e., not as significant. It amounted to 91.3 mln passengers (a decrease from 220.89 mln to 129.56 mln), i.e., 41.3%. Compared to other European countries, Russian airports were in the best position in terms of reduction in passenger turnover (Table 5), since in the former, the decline ranged from 83% (Slovenia) to 60% (Albania).

Table 5.   Decline in annual passenger turnover of Russian airports due to COVID-19 pandemic, passengers (listed in descending order of passenger turnover; airports with passenger turnover in 2019 less than 400 000 people are not included, but those for which passenger turnover in 2020 exceeded this value are included)
Full size table

On average, the decline in passenger turnover at individual Russian airports was 30–48%, or 1.4–1.9 times; at European airports, 2.5–4 times. However, due to a significant decrease in the number of international passengers, the Moscow air hub experienced a larger decline (52%). In 2020, 11 994 thous. passengers of international airlines used the services of its three airports (75.8% less than in 2019); on domestic airlines, the volume of passenger traffic decreased by 29.9%, 36.527 thous. people.

Due to a sharp decrease in the number of international passengers, the maximum reduction in passenger turnover was noted at the airports of Sheremetyevo (60.4%), Vladivostok (58.0%), and Vnukovo (46.7%). A significant drop in passenger turnover (more than 40%) occurred at other large airports (Pulkovo, Yekaterinburg, and Irkutsk), as well as at large airports in the south of the Far East (Vladivostok, Khabarovsk, and Yuzhno-Sakhalinsk).

The passenger turnover of airports in a number of large cities located within the main core of the developed territory of European Russia, with convenient rail links, as well as being in the shadow of larger neighboring airports, significantly decreased: Nizhny Novgorod (48.4%), Voronezh (48.1%), Belgorod (46.0%), Nizhnekamsk (45.2%), and Samara (44.1%). The same thing happened in the south of Western Siberia, where Novosibirsk airport dominates, in the land gravitational zone of which are small airports: Tomsk (53.5%), Barnaul (43.9%), and Kemerovo (42.1%). A significant decline was noted in a number of small airports in the shadow of larger ones: Saransk (55.0%) and Kogalym (48.8%).

Conversely, at main tourist airports (Sochi, Simferopol, and Kaliningrad), where Russians rushed in July–August 2020 instead of the former foreign centers of sea health tourism in the pre-COVID era, conversely, the passenger turnover decreased insignificantly (by only 4–10%). And at the airports of Gelendzhik (+29.7%) and Anapa (+10.5%), it even increased versus 2019 (and in the summer months, at Kaliningrad airport). Growth also occurred in Gorno-Altaysk (+45.0%) and Kursk (+47.4%) due to completion of their reconstruction.

A slight decline (12–29%) was noted in cities far removed from the main area of the continuous developed territory of the country (Petropavlovsk-Kamchatsky, Norilsk, Magadan with no railway passenger traffic, and where air transport remains the only means of communication with Moscow and the rest of the country’s major airports; Murmansk, Novy Urengoy, Noyabrsk, Abakan, Blagoveshchensk, and Chita have rail links, but are significantly removed from the main strip of economic development), as well as airports in the regional centers of the North Caucasus (Makhachkala, Vladikavkaz, and StavropolFootnote 11). A slight decline is also typical of cities that with inconvenient transport–geographical position on the railway network (at a dead end or a connecting branch) within the core of the developed territory of European Russia; therefore, the local airports have not lost a very large number of passengers (Cheboksary, Ulyanovsk, and Izhevsk). Kaluga airport, located in the shadow of the Moscow air hub, where the route network was not closed in spring 2020, has the same type of dynamics.

The minimum decline was noted at the airports of Sochi (3.8%), Makhachkala (6.7%), Saratov (8.7%), Simferopol (9.9%), and Kaliningrad (10.6%).

What are the differences in the passenger turnover dynamics at airports in individual Russian macroregions? Airports of the Moscow Air Hub (MAH), as mentioned above, lost more passengers from the pandemic than others. The group of sea resort airports of Southern Russia were least affected by the negative consequences. Despite the pandemic, passenger traffic at two airports (Gelendzhik and Anapa) increased. At the two largest sea resort airports in Southern Russia (Sochi and Simferopol), the decline was insignificant, while at Mineralnye Vody, it was weak. The passenger turnover of Kaliningrad Airport, which serves the resorts of the Baltic seaside, also experienced an insignificant decline.

Reduced passenger traffic at airports of the Far North was minimal: an insignificant decline for Murmansk airport (13%); weak for Norilsk (16%), Magadan (20%), and Novy Urengoy (28%); moderate for the airports of Arkhangelsk (30%), Syktyvkar (33%), and Yakutsk (39%). The decline at Siberian airports was slightly higher: only in two of them (Chita, Ulan-Ude) was it weak (24–29%); in 9, moderate; and in two, medium (45% (Irkutsk) and 53% (Tomsk)). Here, a moderate decline dominated (from 30% to 44%: Surgut, Novosibirsk, Omsk, Tyumen, Krasnoyarsk, Khanty-Mansiysk, Nizhnevartovsk, Kemerovo, and Barnaul).

Airports of the Far East fell into two groups: weak (Blagoveshchensk, 21%, and Petropavlovsk-Kamchatsky, 27%) and moderate (Yuzhno-Sakhalinsk, 42%; Khabarovsk, 42%; Vladivostok, 58%) declines.

Airports of European Russia (without Moscow, sea resorts, and northern airports) had three types of decline: 3 insignificant (Makhachkala, 7%; Saratov, 9%; Vladikavkaz, 14%); 2 weak (Stavropol, 18%; Astrakhan, 22%); 11 moderate (from 30 to 45%: Rostov-on-Don, Perm, Chelyabinsk, Ufa, Orenburg, Krasnodar, Volgograd, Kazan, Samara, Yekaterinburg, and St. Petersburg), 4 medium (Nizhnekamsk, 45%; Belgorod, 46%; Nizhny Novgorod, 48%; and Voronezh, 48%). A moderate decline dominated at the airports of European Russia.

In general, among the airports in Russia with a passenger turnover of more than 400 000 people, moderate (27 airports) and weak (10) declines prevailed, but there were also airports with insignificant (7) and medium (8) declines. In contrast to European airports, Russian airports had no strong or catastrophic declines in passenger turnover.

Another indicator of the decline is the time lag of the depth of retrodegression (setback (lag) in time)—the number of years that have passed since the moment when the traffic in the past was approximately equal to the value in the COVID crisis year of 2020. Unfortunately, not all airports have annual data on their passenger turnover for the last 25–40 years, which would make it possible to determine what year that was. Nevertheless, the author managed to find such information for a number of European and Russian airports. The retro-lag is indicated in the last column of Table 6 (only a few examples were left in it).

This retrodegression time lag turned out to be the largest (27–40 years) for the major European airports with a large share of international passengers and transit (London–Heathrow, Frankfurt am Main, Hamburg, Paris-Charles de Gaulle, Munich, Amsterdam, Brussels , Zurich, Vienna, Geneva); relatively medium (16–26 years), for medium-sized or large airports with a large share of domestic traffic; relatively short (9–16 years), for the major airports in Russia with a high share of international passengers; short (4–8 years), for almost all other airports in Russia; very short (2–3 years), for resort and tourist airports in Russia and some airports with an increased share of domestic passengers.

It is not possible to determine the time lag (retrodegression) at all airports; therefore, the possibilities of using this indicator for geographical comparisons are very limited.

Table 6. Time lag of depth of retrodegression of passenger turnover at a number of airports in Europe and Russia due to COVID-19 pandemic (passenger turnover, thous. people)
Full size table

DISCUSSION

Torsten Hägerstrand (1967) identified three forms of innovation diffusion within his theory of innovation diffusion (Smirnyagin and Tarkhov, 2013):

(1) hierarchical—the distribution occurs hierarchically through a network of centers and nodes: from the main center, the innovation enters the centers of the second order, from the latter, to centers of the third order, etc., by all ranks of the hierarchy of centers and settlements;

(2) wave (contact)—propagation is frontal in the form of a spatial wave, continually encompassing the entire territory, from one (region) to neighboring territories (regions);

(3) mixed—certain types of innovations spread simultaneously in a hierarchical and wave way.

Using this theory, geographers in the 2000s studied the spatial distribution of influenza and epizootic epidemics (Cliff et al., 2004; Haggett, 2000; Lawson, 2006; Souris, 2019). The geographical features of the spread of these forms of diseases are as follows. Contact (wave) diffusion of infection is characterized by an outbreak in one region (area) and spread to neighboring regions and districts, so that the disease has the highest intensity at the place of origin. Conversely, hierarchical spread is characterized by the onset of disease in a certain place and its jumping to more distant areas and points associated with the initial place of origin via hierarchical connections. Disease diffusion can also be mixed, when wave and hierarchical spreading is observed simultaneously. Whereas in the preaviation era, epidemics spread in linearly-hierarchical manner via land and water transport, in the modern era predominated by long- and medium-range air transport, they diffuse in a pointwise hierarchical manner.

From the viewpoint of the theory of diffusion of innovations, at the early stages, COVID-19 spread exclusively in a hierarchical manner through an extensive air communication system: major and large cities and large urban agglomerations receiving direct flights from Chinese, then Italian air hubs, were the first to suffer. From these, at later stages, infection began to penetrate together with passenger-vectors already through land transport to medium and small cities in the zone of influence of major and large cities; it ended up in the countryside last. Later stages of distribution were characterized by a mixed form of diffusion in which the hierarchical form predominated.

As a result of such a hierarchical spread of the pandemic, primarily large European countries with the largest airports were affected; in the second order, medium-sized countries in the shadow of the zone of influence of the latter; and to a lesser extent, peripheral countries with a more disadvantageous transport and geographic position.

In Europe, airports with a strong decline dominated (65–85%): 172 out of 231. Another 42 airports had moderate declines. Norway stands out, with 10 of 12 airports experiencing a moderate decline. In Italy and Spain, seven airports in each also experienced an average decline. There were almost no cases of weak and insignificant decline in Europe, except for Zaporozhye Airport (weak).

A catastrophic decline (over 85%) was recorded at 12 European airports (out of 231), including 4 British (Cardiff, Exeter, Glasgow–Prestwick, and Newquay), 2 Spanish (Reus and Girona), and 2 Croatian (Pula, Dubrovnik), as well as the German Paderborn, Montenegrin Tivat, and Italian Treviso. Reus (96%), Girona (91%), Pula (89%), Dubrovnik (89%), Paderborn (87%), and Cardiff (87%) had the maximum catastrophic decline.

Eighty-two percent of European airports experienced a severe to moderate decline in passenger turnover as a result of the pandemic. The largest decline in passenger traffic in 2020 occurred at airports in the UK, Spain, Germany and Italy; medium, in Norway; moderate, in Russia.

Of the 285 airports studied (including Russian) for which statistical information on their passenger turnover for 2019 and 2020 was available, 60.4% experienced a strong decline; 17.5%, a medium decline; and 10.5%, a moderate decline. A catastrophic decline was recorded at 12 airports (4.2%); weak, at 11 (3.9%); insignificant, at 7 (2.5%). Passenger traffic increased only at two airports (Anapa and Gelendzhik).

CONCLUSIONS

The COVID-19 pandemic has had a radical impact on air travel and nearly all airports never seen before in postwar Europe.Footnote 12 The study of the spatial differentiation of its consequences on air passenger traffic became possible owing to the use of statistical information on passenger traffic at airports. This latter indicator proved a good quantitative indicator of local, regional, and intercountry differences in such consequences when studying the decline in passenger traffic.

The decline in passenger traffic at almost 300 European airports was extremely spatially uneven. It was strong at the largest airports with a large share of international air transport, medium at medium-sized airports with a minimum share of international traffic, and moderate at Russian airports (due to the rapid restoration of domestic air traffic). The decline and recovery in airport passenger turnover depended directly on the duration of lockdowns and the number of recurring COVID-19 outbreaks, the introduction of entry bans, or the lifting of travel restrictions to certain areas, hotbeds, and countries.

From the COVID crisis, the largest international airports and airports with narrow sea resort and tourist specialization were affected the most; airports that have dominant domestic traffic (including traffic to domestic resorts) and are located in remote and inaccessible areas suffered less. In general, in the entire sampling of 285 airports, strong and medium declines in passenger turnover predominated.

The pandemic has to some extent distorted the territorial structure of air traffic, including the hierarchy of airports throughout Europe as a whole and intranational airport systems. Some of the formerly largest airports in terms of passenger turnover lost their place in the hierarchy in 2020 to other European airports, which bypassed them due to a lower level of decline and rapid recovery of domestic air traffic.