1 Introduction

As a tropical cyclone (TC) prone area, the Bay of Bengal (BoB) frequently suffers from relevant natural disasters and sustains great losses (Mohanty et al. 2015; Chauhan et al. 2018; Kumar et al. 2020). In general, in late pre-monsoon (April and May) and in early post-monsoon (October and November), due to the warm pool phenomenon with other various dynamic processes such as large-scale circulations, air–sea interaction, and vortex-scale convective circulations (Kumar et al. 2021; Reshmi Mohan et al. 2022), about five TCs occur per year over this area (Najar and Salvekar 2010). With further favorable environmental conditions for TCs’ development and intensification, they often move towards land and cause destruction in local (Chauhan et al. 2021b). In fact, the BoB has generated seven severe TCs with individual death tolls in excess of 100,000 (Fritz et al. 2009).

On the one hand, the strong winds of TCs can induce huge waves and storm surges in the nearshore region, which in turn threaten coastal communities and infrastructure, and arouse seawater encroachment to affect ecology (Kundu et al. 2001; Bhaskaran et al. 2020). On the other hand, TCs can directly destroy buildings with their strong wind, bring a heavy precipitation in a short time, and further cause flooding on the land (Chittibabu et al. 2004; Madsen and Jakobsen 2004; Chauhan et al. 2021a). In the Ganges Delta, which has a long and complex coastline with many estuary islands, dense river network, high tidal range, and other adverse factors (Paul et al. 2016), the impact of TC is particularly pronounced in Bangladesh, since its large population density makes it especially vulnerable to high rates of mortality. In fact, since the 1970s, Bangladesh has the bitterest experience with more than 300,000 deaths from a single TC in 1970 (Islam et al. 2021), and about 26% of the Bangladesh total populations are affected by TCs (Cash et al. 2013).

In May 2020, Amphan, a category-5 TC with respect to the Saffir-Simpson scale, formed in the BoB and then struck India (especially West Bengal) and Bangladesh on May 20th successively, making it the costliest storm recorded in the BoB (Priodarshini et al. 2021). The entire variation process of Amphan’s track and its intensity with a 3 h time interval is shown in Fig. 1. It is worth mentioning that Amphan, which was considered the first of its kind after the Odisha Super cyclone 1999 in severity and scale (Kumar et al. 2022), caused less casualties compared to the past super cyclones with similar intensity in this area, e.g., Gorky (1991) and Sidr (2007). Keeping the disaster level as low as possible is certainly related to the improved forecasting, preparedness, and rescue of the local government in recent years (Saha and James 2017; Islam et al. 2021; Chauhan et al. 2021b). In addition, the dynamic characteristics of Amphan itself also deserve attentions. For example, some previous studies focused on the rapid intensifying of Amphan (Reshmi Mohan et al. 2022; Bhate et al. 2023) 2 days before landfall with the specific environmental conditions in the BoB, which also made the intensity of Amphan continuously decreased to category-2 at its landfall, as shown in Fig. 1. Considering that the environment features and the TC impacts are closely correlated, it is important to study in detail the variations of the atmospheric and oceanic parameters and the dynamical characteristics of TC during its life cycle, so that the future TCs in the BoB could be better predicted.

Fig. 1
figure 1

The track and intensity variations of Amphan during 15–21 May 2020 (with time information details). Color dots along the tracks indicate the Saffir–Simpson scale category. The magenta dashed line squares show the locations of two 2° × 2° boxes. The specified districts in Khulna and Barisal Divisions, Bangladesh, with the Amphan’s track around Box 1, are shown as the inset

Full size image

Meanwhile, Amphan occurred at the time of the world-wide ongoing COVID-19 pandemic, which might further deteriorate the situation. When people are affected by natural disasters such as TCs, they are often recommended to temporally move to the relatively closed and crowded public shelters as one of the normal approaches for disaster prevention and mitigation. Nevertheless, such approach should be reconsidered and reassessed under the influence of the COVID-19 pandemic, since maintaining a proper social distance to slow down the spread of the virus under the pandemic becomes a consensus (Baidya et al. 2020; Greenstone and Nigam 2020; Kumar et al. 2021). Many investigations have pointed out that the risk perception of coastal people about the potential infection threats of the COVID-19 pandemic can influence evacuation decisions during Amphan (Alam and Chakraborty 2021; Hadi et al. 2021; Alam et al. 2022). From another perspective, after the nationwide lockdown, the destruction of infrastructure and the pollution of clean water caused by Amphan could also further increase the difficulty of epidemic prevention and disaster recovery (Pramanik et al. 2022). The possible correlation between the short-term TC disaster and the long-term COVID-19 pandemic is still unclear. Amphan provides a valuable opportunity to scrutinize the above relationship, being useful to draw lessons and insights for effective disaster prevention and mitigation when multi-type disasters coexist.

The objective of the present study is to document the dynamic characteristics of Amphan and its aftermaths, and to conduct the detailed investigation of the observation data and reanalysis data to study the impact of Amphan on atmospheric, meteorological, oceanographic, and hydrodynamic parameters in the entire process from its formation to landing, as well as the final cyclolysis. Finally, in light of the local development of the COVID-19 pandemic, the measures in the face of multi-type disasters, such as TCs and epidemic, are also within the scope of the present study.

2 Data and methodology

TC data used in this study are derived from the International Best Track Archive for Climate Stewardship (IBTrACS) v04 (https://www.ncei.noaa.gov/products/international-best-track-archive). IBTrACS is the most complete global collection of TCs available and merges best track data from multiple agencies in the world (Knapp et al. 2010). To keep the data quality as homogeneous as possible, only the IBTrACS data provided by USA agencies were used. In the analysis of sea surface temperature (SST) anomaly, data with a horizontal resolution of 0.25° × 0.25° from the Optimum Interpolated Sea Surface Temperature datasets (https://www.ncei.noaa.gov/products/optimum-interpolation-sst) were used, which incorporates observations from different platforms, such as satellites, ships, and buoys (Reynolds et al. 2007; Huang et al. 2020).

Historical TC-related environment data, such as precipitation, wind field (horizontal wind and omega), and relative humidity in the BoB, are available from the fifth generation of ECMWF atmospheric reanalysis of the global climate and weather (ERA5) (Hersbach et al. 2019a, b). These parameters are all available at a spatial resolution of 0.25° × 0.25° (https://cds.climate.copernicus.eu/cdsapp#!/search?type=dataset).

Meanwhile, several observation data were selected to show the hydrodynamic characteristics of Amphan. The in-situ ocean subsurface time-series measurements are from the Ocean Moored buoy Network for the Northern Indian Ocean (OMNI) buoys (https://incois.gov.in/portal/datainfo/mb.jsp). The detailed sampling information, data resolution, and sensor types used in the OMNI buoy system are described in Venkatesan et al. (2013).

Based on the 1-min sea-level data from the tidal gauge station at Chittagong (http://www.ioc-sealevelmonitoring.org/station.php?code=chtt) provided by the sea-level station monitoring facility which is hosted by Flanders Marine Institute and Intergovernmental Oceanographic Commission (2022), the temporal variation of Amphan-induced storm surge at Chittagong was studied. To accurately distinguish the components, the term “surge” refers to the meteorological contribution to the increase of sea-level, and the term “residual”, which is calculated by subtracting the astronomical tide from the recorded sea-level data, contains surge, tide–surge interaction, and harmonic prediction errors (Horsburgh and Wilson 2007). Here, the T_Tide harmonic analysis toolbox in MATLAB (Pawlowicz et al. 2002) was used to extract the predicted astronomical tide.

To fully document the characteristics of the surrounding environmental field during the life cycle of Amphan, the entire BoB was chosen as the target region of the present study. Times mentioned in this study are all in GMT if there is no further specification. According to the Amphan’s track and the relative distance among the buoys, data recorded from the OMNI buoys BD08 (18.2° N, 89.7° E) and BD11 (13.5° N, 84.0° E) are used to analyze the development process of Amphan. Locations of BD08 and BD11 are specified in Fig. 1.

Following Chauhan et al. (2020, 2021a), two 2° × 2° land boxes are considered here: one covering Bangladesh (Box 1) and the other covering West Bengal (Box 2). As shown in Fig. 1, Amphan’s track is located in between Box 1 and Box 2, whose boundary (west, south, east, and north) in degree is (89° E, 22.75° N, 91° E, 24.75° N) and (86.75° E, 22° N, 88.75° E, 24° N), respectively. These two boxes are selected to analyze the variations in environmental parameters before and after landing. In addition, to intuitively reflect the influence of Amphan in local, the analyses of various parameters around the location of landfall (i.e., Bakkhali, West Bengal) are also conducted.

3 Spatiotemporal characteristics of Amphan-related parameters

3.1 An overview of Amphan

Recorded since May 16th in 2020, Amphan rapidly developed and reached its peak intensity in the ocean at 12:00 on May 18th. Afterwards, Amphan’s intensity gradually weakened during the northward movement until it landed and dissipated. With the maximum 1-min maximum sustained wind (MSW) speed of about 74 m/s and the minimum sea-level pressure (MSLP) of approximately 901 hPa during its entire life cycle, Amphan is the strongest TC to strike the Ganges Delta since the 2007 Sidr.

Amphan caused certain damages to human communities along its track. According to the data obtained from the Bangladesh Water Development Board (BWDB, available online at https://www.bwdb.gov.bd/project-report/#page-1), Table 1 summarizes the damages to various structures (e.g., embankment, river bank, and other infrastructures) caused by Amphan-induced floods in some coastal districts in Bangladesh. As the first district affected by Amphan, Satkhira, just north of the Sundarbans Mangrove Forest, was the most severely damaged district by floods in Amphan with a total embankment damage length of 111.43 km, river bank erosion length of 0.2 km, and embankment maintenance works damage length of 8.73 km. In Barisal Division, which is relatively far away from Amphan’s track, although situations were not as serious as in Satkhira, the damages to the exposed coast, e.g., Barguna, Patuakhali, Pirojpur, and Bhola, were still relatively significant. Amphan-induced floods and storm surges damaged a large number of embankments, roads, and houses, which led to the inundation of numerous villages and even some cyclone shelters (Hassan et al. 2020; Halder et al. 2021). In addition, the destruction of transport infrastructure may further delay the post-disaster relief work and prolong the disaster period.

Table 1 Summary of damages to various structures in some districts in Khulna and Barisal Divisions caused by Amphan from BWDB
Full size table

Considering that there have been some studies that specifically describe the movement of Amphan from formation to landing (Roman-Stork and Subrahmanyam 2020; Vissa et al. 2021; Chauhan et al. 2021b), this section will mainly focus on analyzing the variations of the atmospheric, meteorological, and oceanographic parameters during Amphan’s life cycle.

3.2 SST anomaly and relatively humidity

As shown in Fig. 2, on May 14th, before the formation of Amphan, there was a significant SST anomaly in the BoB, with an extreme value of nearly 3 °C, which was greater than the SST anomaly in the same period of the previous year, i.e., 2019 (Chauhan et al. 2020). It is for this reason that such a large-scale warm pool provides a huge amount of energy for the development of Amphan (Naskar and Naskar 2021), making it far more intense than the cyclone Fani in May 2019. With the further development of Amphan and its northward movement, due to the intensification of the cyclone circulation and the enhancement of wind stress, the upper ocean was violently mixed, and the negative SST anomaly gradually appeared in the center of the BoB (D’Asaro et al. 2014; Chauhan et al. 2018). Nevertheless, SST cooling would in turn reduce the enthalpy supply to a TC, limiting its intensity development (Price 1981; Guan et al. 2021). This also caused Amphan’s intensity to weaken rapidly after May 18th, greatly reducing the threat it posed when landed on May 20th (Chauhan et al. 2021b). At the same time, the positive SST anomaly that still existed along the northern coast of the BoB generally disappeared after Amphan passed on May 22nd.

Fig. 2
figure 2

Spatial–temporal variations of the daily SST anomaly over the BoB during Amphan. The magenta markers show the locations of some major cities in Bangladesh, including Dhaka (DK), Rajshahi (RS), Sylhet (SH), Chittagong (CG), and Khulna (KN). The black line indicates the track of Amphan

Full size image

Over the same period, Fig. 3 shows the changes in relative humidity over the BoB. Due to the strong interaction between Amphan and the high-temperature sea water and the increased moisture-holding capacity of warmer air over the oceans (Chadwick et al. 2016), it is observed that the high relative humidity zone is moving continuously towards land around the center of Amphan. After Amphan made landfall, the high relative humidity air mass along the Amphan track became a source of heavy rainfall.

Fig. 3
figure 3

Spatial–temporal variations of the daily relative humidity at 700 hPa pressure level over the BoB during Amphan. The black line indicates the track of Amphan

Full size image

3.3 Wind field

Figure 4 shows the change of the horizontal wind field at 850 hPa pressure level over the BoB during the Amphan. With the development of Amphan, the anti-clockwise wind circulation pattern gradually became clear. Before landfall, the wind speed close to the center could reach 30 m/s or more. Nevertheless, on the day of landfall (May 20th), the wind speed decreased, and the anti-clockwise wind pattern generally disappeared on May 22nd. It should be noted that due to the accuracy of spatiotemporal resolution, the wind speed of Amphan specified in ERA5 is smaller than the MSW provided by meteorological agencies (e.g., IBTrACS data). In addition, it is found that the high wind speed zone around the Amphan center highly overlapped with the large relative humidity zone shown in Fig. 3, indicating that the Amphan wind field carried a lot of water vapor to the land during its northward movement.

Fig. 4
figure 4

Spatial–temporal variations of the daily wind speed and direction at 850 hPa pressure level over the BoB during Amphan. The black line indicates the track of Amphan

Full size image

Regarding Fig. 5, it is noted that the pattern of wind field at 10 m height (10 m wind) is similar to that of wind at 850 hPa pressure level, while the range and value of the high wind speed area of the former are obviously smaller than those of the latter. In addition, far from the Amphan center, the directions of the two wind fields at different altitude levels do not appear to be so consistent. In particular, due to the different surface properties between land and sea, the 10 m wind caused by Amphan on the land was weaker than that over the sea at the same time (especially on May 20th), which could further affect the asymmetric structure of the TC core (Wong and Chan 2006).

Fig. 5
figure 5

Spatial–temporal variations of the daily wind speed and direction at 10 m height over the BoB during Amphan. The black line indicates the track of Amphan

Full size image

Figure 6 shows the changes in omega at 850 hPa pressure level during Amphan. Before the formation of Amphan (e.g., May 14th), only a small part of the entire ocean existed negative omega values. With the development of Amphan, the zone with large omega negative values expanded rapidly. Due to the large positive SST anomaly in the case of Amphan, as shown in Fig. 2, the corresponding increment in vertical motion was produced (Reshmi Mohan et al. 2022). The weak vertical shear and the atmospheric instability caused by the diabatic heating also enhanced the convection and vertical updraft (Naskar and Naskar 2021; Bhate et al. 2023). Combined with Fig. 3, it is confirmed that an upward movement zone of air mass also occurred along with the high wind speed and high relative humidity zone as above-mentioned. Especially on the day of landfall, the large negative omega values (up to − 1.5 Pa/s) led to air convection with high relative humidity, which in turn triggered intensive precipitation on the land, resulting in local floods and damages in Bangladesh and West Bengal (Chauhan et al. 2021b). At the same time, there was a downdraft zone in the west outer core of Amphan. After the cyclolysis of Amphan (e.g., May 22nd), the omega pattern over the BoB, similar to those of other wind fields, had also roughly restored to the state before Amphan.

Fig. 6
figure 6

Spatial–temporal variations of the daily omega at 850 hPa pressure level over the BoB during Amphan. The black line indicates the track of Amphan

Full size image

Figures 7 and 8 show the changes of the relative vorticity at 850 hPa pressure level and the divergence at 200 hPa pressure level over the BoB during Amphan, respectively. For a specific TC case, air usually converges in the lower troposphere, rises, and diverges in the upper troposphere. Consistent with the distribution and evolution of the wind field, the relative vorticity and divergence field exhibited positive values in the vicinity of the Amphan center. Germinating before the officially recorded formation of Amphan, the center of positive anomalies developed as Amphan intensified offshore, weakened as it approached the land, and finally vanished as it dissipated.

Fig. 7
figure 7

Spatial–temporal variations of the relative vorticity at 850 hPa pressure level over the BoB during Amphan. The black line indicates the track of Amphan

Full size image
Fig. 8
figure 8

Spatial–temporal variations of the divergence at 200 hPa pressure level over the BoB during Amphan. The black line indicates the track of Amphan

Full size image

3.4 Temporal variations of various parameters

Figure 9 shows the temporal variations of some meteorological parameters provided by ERA5 over Boxes 1 and 2 and the landfall location of Amphan. The dynamical characteristics over Boxes 1 and 2 are very similar in the process of Amphan’s movement. In general, all parameters rapidly returned to their pre-Amphan states after its cyclolysis. Figure 9a shows that the water vapor over the coastal areas reached a high level on the day before Amphan landfall, which is consistent with the characteristics reflected in Fig. 3. Accordingly, there were obvious precipitations in above-mentioned areas, especially two peaks of hourly precipitation around the landing point as shown in Fig. 9b. When daily precipitation is considered, the average precipitation in Box 2 could reach 90 mm on May 20th, while the large amount of precipitation with short duration and associated landslide and flood could lead to more threats (Polemio and Lonigro 2015).

Fig. 9
figure 9

Temporal variations of a relative humidity (700-hPa), b hourly precipitation, c 10 m wind speed, d 850-hPa wind speed, e omega (850-hPa), f relative vorticity (850-hPa), and g divergence (200-hPa) over Boxes 1 and 2 and the landfall location during Amphan. From left to right, three black dotted lines correspond to the time instants when Amphan was first recorded, landing, and last recorded in the IBTrACS dataset, respectively

Full size image

Regarding the variations of the horizontal wind fields in Fig. 9c and d, it is found that since May 18th, both the local 10 m wind and 850 hPa wind began to increase continuously. Especially near the landing point, the average wind near the ground could reach about 25 m/s when Amphan made landfall, which posed a serious threat to local residents and various infrastructures. Figure 9e shows the intensity of air motion in the vertical direction. After May 19th, the ascent (negative values of omega) of the local air gradually became significant, and reached its extreme value at the time of landfall on May 20th, and then quickly returned back to the normal level, indicating that the TC center had passed through this area. At the same time, relative vorticity also appeared a temporal variation consistent with the omega, as shown in Fig. 9f. Nevertheless, the values of divergence were fluctuating around zero since May 17th, and there was a significant reversal of its extreme values (from negative to positive) before the Amphan’s landfall in Fig. 9g.

3.5 Buoy data

Figure 10a–c shows the temporal variations of typical meteorological parameters recorded by BD08 and BD11, including air pressure, wind speed, and wind direction. There are intermittent missing data in the recording as indicated by the discontinuous results in Fig. 10. Considering Figs. 1 and 10a, the gradual decrease of air pressure is associated with the passage of Amphan. The BD11 data show an earlier recorded minimum pressure of 991.46 hPa at 11:00 on May 18th, whereas the minimum pressure of 987.65 hPa occurred later at 12:00 on May 19th for BD08 data. Because both buoys are far away from the track of Amphan, the recorded air pressure data are obviously higher than the corresponding pressures provided by IBTrACS. In Fig. 10b, a significant increase in wind speed is observed at BD08, which reached a maximum of 19.84 m/s at 18:00 on May 19th, about 18 h before the Amphan landing. Nevertheless, the wind speed information at BD11 was missing during Amphan. As for the wind direction in Fig. 10c, it gradually changes from the southeasterly (i.e., 90–180°) to the southwesterly (i.e., 180–270°) at BD08 with the passage of Amphan, indicating a clockwise rotation of wind direction, which is in accordance with the general scene for the variation of local wind direction during a TC event in the northern hemisphere (Price 1981), since BD08 is located on the right side of the TC track. In contrast, the wind direction at BD11, located on the left side of the Amphan track, was northeasterly (i.e., 0–90°) initially, which changed to the northwesterly (i.e., 270–360°) when Amphan passed by, revealing an anti-clockwise rotation feature during the entire process. After Amphan made landfall, its impact on the sea area where the moored buoys are located significantly weakened, and the meteorological parameters recorded by these two buoys tended to be equal, as shown in Fig. 10a–c. In fact, regarding the time variations of 10 m wind parameters, similar features could also be observed in Fig. 5, indicating the reliability of the measurement results observed by moored buoys.

Fig. 10
figure 10

Temporal variations of the a air pressure, b wind speed, c wind direction, d current speed, e current direction, f significant wave height, g mean wave direction, and h mean wave period recorded at buoys BD08 (3 h recording interval) and BD11 (1 h recording interval) during Amphan. From left to right, three black dotted lines correspond to the time instants when Amphan was first recorded, landing, and last recorded in the IBTrACS dataset, respectively

Full size image

Figure 10d–h shows the temporal variations of typical hydrodynamic parameters measured by BD08 and BD11, including current speed, current direction, wave height, wave direction, and wave period. In Fig. 10d, the current speed recorded at BD11 reached its maximum value of 1.34 m/s at 17:00 on May 18th. Before the above maximum current speed, a second peak occurred at BD11 on May 17th. As for BD08, there was no significant change in current speed till 12:00 on May 19th, after which a significant increase occurred with a maximum value of 1.56 m/s recorded at 9:00 on May 20th. Subsequently, a second peak appeared on May 21st after the landing of Amphan. The peak current speed at BD08 is larger than that at BD11. From the time-history of Amphan’s track in Fig. 1 and the wind speed in Fig. 10b, it could be confirmed that there was an obvious time lag regarding the occurrence of the peak current speed. This is ascribed to the fact that the current speed is driven by (thus lags) the wind during the TC period (Venkatesan et al. 2014). As shown in Fig. 10e, the current direction at BD11 gradually changes from the southwest (i.e., 180–270°) to the southeast (i.e., 90–180°) with a slight anti-clockwise rotation during Amphan, whereas the current direction at BD08 exhibited a clear clockwise rotation during this process. This is consistent with the rotation of the wind speed, as shown in Fig. 10c.

The wave parameters also changed associated with the propagation of Amphan. Figure 10f illustrates that the sea surface was relatively calm before Amphan and the increase in significant wave height was obvious during Amphan. The maximum significant wave height at BD11 is 6.25 m occurring at 10:00 on May 18th, which is smaller than the corresponding maximum value at BD08, being 8.41 m occurring at 18:00 on May 19th. BD08 is located on the right side of the Amphan track, where higher wind speed prevails due to the wind field asymmetry which could generate a relatively larger wave height at BD08. In addition, the fetch for wave generation is extended on the right side of the TC center, since waves at BD08 tend to move forward with the Amphan movement and therefore remain in the high wind region for an extended period. Nevertheless, waves generated at BD11 (left side of the Amphan track) propagate in the opposite direction to the Amphan movement and rapidly move away from the high wind region. This is in agreement with Young and Burchell (1996). Figure 10g shows the temporal variation of wave direction, which is defined as the direction from which the waves come, being the same as the conventional definition of wind direction. When Amphan passed by, the wave direction recorded at BD11 was from northeasterly to northerly resulting in an anti-clockwise rotation. The wave direction recorded at BD08 was from southerly to southwesterly in a clockwise manner, which, however, was relatively unobvious. Figure 10g also shows that there was a periodic oscillation in the wave direction at BD08 after Amphan landing. As for the wave periods recorded at these two buoys, both increased when Amphan approached and then decreased after it made landfall, as shown in Fig. 10h.

3.6 Storm surge

Due to the wide continental shelf with shallow bathymetry and the funneling shape of the BoB, TCs are prone to cause high storm surges, which further lead to tremendous losses of life and property (Hussain and Tajima 2017). As shown in Fig. 11, temporal variation of residual at Chittagong shows periodical fluctuations owing to the tide–surge interaction, which as a non-linear process depends on the amplitude and phase of the tidal constituents and the storm (McLaughlin et al. 2022) being significant in shallow seas and estuaries (Spicer et al. 2019; Antony et al. 2020). After removing the high-frequency components with a low-pass filter (Feng et al. 2016), temporal variation of the surge signal at Chittagong is obtained, which began to rise rapidly on May 19th, reached its peak value about 1 m around 18:00 on May 20th (after Amphan landing), and then decreased gradually. In general, the Amphan-induced storm surge at Chittagong was not large though Amphan reached a category-5 intensity during its lifetime. One possible reason is that Amphan reached its lowest MSLP (the largest MSW speed) in the offshore region, being very much early before its landfall, and its intensity significantly reduced since May 19th as indicated in Fig. 1. In addition, Chittagong is located far away from the Amphan track, which might also be the reason for the relatively small surge height.

Fig. 11
figure 11

Temporal variations of the Amphan-induced storm surge at Chittagong, Bangladesh

Full size image

4 Possible correlation between Amphan and COVID-19 pandemic

It is evident that COVID-19 could spread readily by asymptomatic (Oran and Topol 2020) and airborne transmission (Morawska and Milton 2020). Therefore, during the pandemic period when people need to maintain social distancing, Amphan might deteriorate the epidemic prevention effort in Bangladesh and West Bengal if many TC refugees crowded in the relatively confined shelters during the disaster evacuation. To reveal the possible relationship between Amphan and COVID-19 pandemic, a simple statistical analysis on the number of daily infected people before and after the life cycle of Amphan was conducted for a preliminary investigation. According to the global COVID-19 data repository (available online at https://github.com/CSSEGISandData/COVID-19) provided by Dong et al. (2020) and the bulletins reported by the government of West Bengal (available online at https://www.wbhealth.gov.in/pages/corona/bulletin), as well as the data dashboard updated by the Ministry of Health and Family Welfare of Bangladesh (http://103.247.238.92/webportal/pages/covid19.php), the daily data of newly diagnosed cases could be accessed, and correspondingly, the first derivative with respect to time could also be calculated. The corresponding results are shown in Fig. 12. The life cycle of Amphan and the incubation period of the virus are taken as the important time nodes to analyze the possible correlation between Amphan and COVID-19 pandemic. According to Lauer et al. (2020), the incubation period of COVID-19 was set as 14 days in this study.

Fig. 12
figure 12

Daily variation of the newly confirmed COVID-19 cases in a Bangladesh, b West Bengal, India, and c Khulna Division, Bangladesh. From left to right, three black dotted lines correspond to the time instants of Amphan’s first and last records in the IBTrACS dataset, and the 14-day incubation period after the last record of Amphan, respectively

Full size image

After curve-fitting the daily diagnosed number of infected people in Bangladesh, it is found that the temporal variation of the increasing rate of the diagnosed number (the first derivation of the fitted curve) presents a local peak within the 14-day incubation period after Amphan, as shown in Fig. 12a. Regarding the situation of West Bengal in Fig. 12b, the corresponding local peak of the increasing rate also occurs within the 14-day incubation period after Amphan, the same as that in Bangladesh. Similarly, Kumar et al. (2021) and Kumar et al. (2022) also noted the increase in local confirmed COVID-19 cases after the passage of Amphan. This might be ascribed to the massive evacuation efforts in the relatively concentrated places during Amphan, e.g., shelters. In Khulna Division of Bangladesh (right beside the Amphan’s track as shown in Fig. 1) where the most severe damages occurred during Amphan, the number of newly confirmed cases increased rapidly after Amphan landing which, however, was almost nil before the event, as shown in Fig. 12c. According to the post-Amphan situations in local (Hassan et al. 2020; Halder et al. 2021), most of the people neither maintained social distancing nor wore face masks when they had to take refuge or act together against Amphan-induced disasters. Such direct face-to-face communication behaviors could increase the risk of virus transmission. Though further comprehensive analyses are definitely needed, Fig. 12, to a certain extent, suggests that there could be a possible correlation between the Amphan event and the intensive outbreak of the local COVID-19.

Shultz et al. (2020) pointed out that reliable approaches for population protection from TCs often tend to gather people together into safe shelters. Nevertheless, strategies aimed at curbing the spread of COVID-19 rely on physical distancing and keeping people apart. This incompatibility undoubtedly results in a dilemma. Hence, further attentions need to be paid in the future for the situation under which multi-type disasters, requiring different (or even conflicting) disaster prevention and mitigation measures, occur simultaneously.

5 Conclusions

In May 2020, a powerful and catastrophic TC, Amphan, formed in the BoB and attacked India and Bangladesh successively. In this study, with comprehensive data from multiple sources, we documented the dynamical characteristics of Amphan and its aftermaths on atmospheric, meteorological, oceanographic, and hydrodynamic parameters in its entire life cycle. Meanwhile, considering the fact that Amphan occurred in accompany with the ongoing COVID-19 pandemic, the possible correlation between them was also studied. Main conclusions are summarized as follows:

  1. 1.

    With the abundant energy provided by a wide range of high SST anomaly, Amphan developed rapidly and caused significant variations in the wind field and water vapor transport, which ultimately led to the local heavy precipitation and strong winds in the coastal areas during its passage.

  2. 2.

    Amphan’s impacts on oceanographic and hydrodynamic parameters (e.g., ocean waves, currents, and surges) are evaluated, which is helpful for the better prediction of the future TCs.

  3. 3.

    It is confirmed that the number of local newly infectious COVID-19 cases increased after Amphan, especially in Khulna Division, Bangladesh, indicating that the outbreak of the local COVID-19 could possibly be intensified by Amphan.

  4. 4.

    In the light of the fact that abiding the epidemic prevention regulations is hard to be implemented during the TC event due to the massive evacuation and post-disaster reconstruction efforts, more attentions should be paid when dealing with the prevention and mitigation efforts for the multi-type and coexisting disasters if different or even conflicting measures are required.