Figure
1.
Extents of the Arctic Ocean and its marginal sea areas (red box shows the location of the Fram Strait) (a), and zonal gate and meridional gate of the Fram Strait (b).
| Citation: | Tiehan Liao, Haigang Zhan, Xing Wei, Weikang Zhan. Impacts of human activities on morphological evolution in the Modaomen Estuary, China[J]. Acta Oceanologica Sinica, 2023, 42(5): 79-92. doi: 10.1007/s13131-022-2064-7
|
The sea ice in the Arctic Ocean has experienced a dramatic decrease in the past 40 years (Comiso et al., 2017; IPCC, 2019), and it will continue to decrease (Serreze and Meier, 2019). Multi-year ice has greatly reduced, and the loss of multi-year ice in the Arctic Ocean cannot be replenished, making the first-year ice dominant (Kwok and Untersteiner, 2011), and the extent of multi-year ice has retreated nearly 30% to the north (Ke et al., 2013). The Arctic will probably be ice-free in September prior to the end of the twenty-first century (Boé et al., 2009; Screen and Deser, 2019). Sea ice reduction will change the surface albedo and increase the absorption of solar radiation (Perovich et al., 2007), which in turn will promote the further melting of sea ice and form a positive feedback effect; sea ice reduction will also affect the distributions of fresh water and phytoplankton (Serreze et al., 2007; Fujiwara et al., 2014), changes in carbon flux, heat and the climate at the middle and low latitudes (Butterworth and Miller, 2016; Wu et al., 2014). The decline in sea ice can be attributed to two reasons: one is the increase in the melting of sea ice due to the influence of thermodynamics, and the other is the outflow of sea ice in the Arctic Ocean (Zhang et al., 2000; Zwally and Gloersen, 2008). The Fram Strait plays an important role in Arctic sea ice loss (Jung and Hilmer, 2001; Bi et al., 2016). This strait acts as the main export channel for sea ice in the Arctic Ocean (Smedsrud et al., 2011) and plays a significant role in seawater and material exchanges between the Arctic Ocean and the North Atlantic. On average, sea ice equivalent to 10% of the Arctic Ocean area flows out of the Fram Strait every year (Kwok and Untersteiner, 2011), and the transport of sea ice from the Fram Strait to the Norwegian Sea accounts for 25% of the total fresh water exported out of the Arctic (Serreze et al., 2006).
Early researches on sea ice export relies mainly on field buoy observations or upward-looking sonar (ULS) instruments, obtains sea ice information (sea ice velocity, thickness and concentration) of a whole channel through interpolation or modelling (Bi et al., 2018b), and then estimates the exported sea ice area and volume. Buoy observations are important observational data for the study of Arctic sea ice outflow. Lei et al. (2016) utilized buoy data (42 buoys) to obtain the variations in ice kinematics of the Arctic outflow region; the meridional ice velocity dramatically accelerated when ice drifted southward into the Fram Strait. Moritz (1988) utilized in-site data and estimated the average annual area flux (−840×103 km2) from 1979 to 1984. Based on the relationships between the sea level pressure gradients on the eastern and western sides of the Fram Strait and the sea ice velocity, Vinje et al. (1998) used moored sonars data to estimate the annual ice area export (−1100×103 km2) and the annual ice volume export (−2846 km3) from 1990 to 1996. However, it is difficult to obtain large-scale and continuous sea ice information due to the small number and short-term field observations. The development of remote sensing has provided conditions for the study of large-scale sea ice change. Many scholars use satellite observations to study Arctic sea ice velocity and export (Spreen et al., 2009; Babb et al., 2013).
Kwok and Rothrock (1999) obtained the exported sea ice volume from 1990 to 1995 based on ULS thickness and passive microwave sea ice motion data. They concluded that the average volume flux from October to May contributed more than 76% of the total annual volume flux. Kwok et al. (2004) showed that the decadal net exported area of sea ice in the 1990s was larger than that in the 1980s, and the difference was approximately half the annual average (approximately 400×103 km2). Kwok (2009) used microwave radiometer data to study the exported sea ice area through the Fram Strait from 1979 to 2007 and showed that the ice area export flux in winter was approximately seven times that in summer. Based on ice drift data derived from Advanced Synthetic Aperture Rada, Smedsrud et al. (2017) obtained the sea ice area export through the Fram Strait from 2004 to 2010 and found that the recent ice area export was approximately 25% larger than that during the 1960s due to stronger geostrophic winds. This may be an important factor that has resulted in the dramatic decrease in sea ice during recent decades. Bi et al. (2018a) studied sea ice export from 2011 to 2015 and found that the mean winter accumulative volume flux from 2011/2012 to 2014/2015 was lower than that from 1990/1991 to 1993/1994 and from 2003/2004 to 2007/2008, which was mainly due to sea ice motion changes. Ricker et al. (2018) showed that the ice volume export across the Fram Strait was a main controlling factor of multi-year ice volume changes in the Arctic. However, due to the limitation of remote sensing data, satellite data are generally used to obtain the winter sea ice export, and the understanding of summer sea ice export is not enough.
Some scholars have used physical-based models to simulate sea ice parameters, such as the Arctic sea ice mass balance model (Thomas et al., 1996) and the coupled general circulation model (Jung and Hilmer, 2001). Zhang et al. (2017) calculated the sea ice volume export through the Fram Strait from 1979 to 2012 by using sea ice thickness data simulated by the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) and found an extremely large ice volume export between the late 1980s and mid-1990s. The Arctic Dipole (AD) plays important roles in the variation in ice volume export through the Fram Strait (Zhang et al., 2017). Min et al. (2019) estimated the ice volume export during the melt season (from May to September) and freezing season (from October to April) from 2010 to 2016 based on the combined model and satellite sea ice thickness data set, which assimilated CryoSat-2 (CS2) data and Soil Moisture and Ocean Salinity (SMOS) thickness data together with sea ice concentration (SIC) observations. Their results showed that the ice volume export during the melt season was approximately 50% of that during the freezing season.
To date, many studies have been performed on the sea ice export flux through the Fram Strait, especially in winter (Hilmer and Jung, 2000; Bi et al., 2018b). These researches focus on estimating the sea ice area export and volume export using remote sensing data or modelling data and analysing the influences of the Arctic Oscillation (AO) and North Atlantic Oscillation (NAO) on sea ice export. Based on sea ice thickness and velocity data provided by the Copernicus Marine Environment Monitoring Service (CMEMS) and SIC data from the National Snow and Ice Data Center (NSIDC), this study estimates the sea ice volume export through the Fram Strait during 2011–2018, studies the contributions of the sea ice thickness, sea ice velocity and SIC to the sea ice volume export separately, and analyses the relationships between the sea ice volume export and the AO, NAO, and AD. In addition, we explore the impact of Ekman transport (ET) on the sea ice volume export.
The Arctic Ocean is connected to the Bering Sea through the Bering Strait and connected to the Greenland Sea through the Fram Strait (Fig. 1a). The Fram Strait is located between Greenland and Svalbard and acts as the channel between the two, thereby playing an important role in ice transportation from the Arctic Ocean to North Atlantic. The Fram Strait export gate is divided into the zonal gate and meridional gate for calculating the sea ice volume export through the Fram Strait. We define the zonal gate and meridional gate as shown in Fig. 1b according to Krumpen et al. (2016) and Ricker et al. (2018). The zonal gate is located at 82°N between 22°W and 20°E, and the meridional gate is located at 20°E between 80.5°N and 82°N. This gate location is chosen for better comparison to prior studies (Min et al., 2019).
The sea ice thickness and velocity are accessible through the CMEMS, which offers reanalysis data in the Arctic from 1991 to 2018 (Product identifier: ARCTIC_REANALYSIS_PHY_002_003). The reanalysis data is produced from the Towards an Operational Prediction system for the North Atlantic European coastal Zones (TOPAZ4). TOPAZ4 is developed based on a coupled ocean and sea-ice assimilation system. The hybrid coordinate ocean model, which includes 28 hybrid layers in vertical, is utilized, coupled with a sea ice model (CMEMS, 2020; Xiu et al., 2021). The Ensemble Kalman Filter is adopted to assimilate remote sensing data into the numerical model to obtain the final data set (CMEMS, 2020). The sea ice velocities which are used as assimilation data are provided by the French Research Institute for Exploitation of the Sea/Laboratory of Oceanography from Space and the Ocean and Sea Ice Satellite Application Facility (CMEMS, 2020). Since 2014, the sea ice thickness has assimilated data from the CS2 altimeter and SMOS radiometer (CMEMS, 2020). TOPAZ4 ensures that sea ice data also exist in summer.
The TOPAZ4 ice thickness has been evaluated (Xiu et al., 2021) using satellite-derived ice thickness (CS2SMOS), modeled ice thickness (PIOMAS; the Combined Model and Satellite Thickness (CMST)) and field observations. TOPAZ4 ice thicknesses agree well with CS2SMOS, PIOMAS and CMST ice thicknesses, and large differences mainly occurred in the Canadian Arctic Archipelago in September. Although TOPAZ4 overestimates the sea ice in the Beaufort Sea and underestimates the sea ice in the north of the Canadian Arctic Archipelago in March and April, the error is acceptable. Based on the ULS ice thickness, the bias of TOPAZ4 ice thickness is 0.11 m. The maximum mean bias is 0.33 m in summer and the corresponding RMSE is 0.62 m. The mean bias is within ±0.2 m in winter, spring and autumn, and RMSE is less than 0.52 m. TOPAZ4 ice velocity has been evaluated (CMEMS, 2020) using the International Arctic Buoy Program buoy data. TOPAZ4 ice velocity is generally higher than the buoy velocity, and variations in ice velocity are consistent with variations in buoy velocity. The mean bias of TOPAZ4 ice velocity is 2.49 km/d and the corresponding RMSE is 5.08 km/d. TOPAZ4 ice velocity in the Robeson Channel has been assessed using Sentinel-1 images (Liu et al., 2022). TOPAZ4 underestimates sea ice velocity in the Robeson Channel with an mean absolute error of 13.36 km/d. We select daily TOPAZ4 sea ice thickness and velocity data during 2011–2018 with a spatial resolution of 12.5 km×12.5 km.
The SIC comes from the NSIDC. We choose two products that are obtained from several satellite sensors, including the Scanning Multichannel Microwave Radiometer (SMMR), Special Sensor Microwave/Imager (SSM/I) and Special Sensor Microwave Imager/Sounder (SSMIS). The data are Nimbus-7 SMMR DMSP SSM/I-SSMIS passive microwave data (NSIDC-0051) and near-real-time DMSP SSMIS daily polar gridded sea ice concentrations (NSIDC-0081), which provide the SIC from October 1978 to the present. NSIDC-0081 is generated to fill the data gap of NSIDC-0051. Cavalieri et al. (1992) evaluated the SIC, the mean bias during the growth season is within ±3%, and the standard deviation is 4%. Daily data covering the period from 2011 to 2018 are selected with a spatial resolution of 25 km×25 km.
European Center for Medium-Range Weather Forecast (ECMWF) Re-Analysis 5 (ERA5) is the fifth generation ECMWF reanalysis dataset. ERA5 began to provide data to the public on June 14, 2018, replacing the ERA-Interim reanalysis data. ERA5 uses data assimilation to combine model data with various observation data. It has provided a large number of estimates of atmospheric, ocean, and land parameters from 1979 to the present. Monthly 10 m u-component of wind (u10, m/s) and 10 m v-component of wind (v10, m/s) data are used to calculate ET. Monthly mean sea-level pressure (MSLP, Pa) data are downloaded to calculate the AD index. The downloaded data has a spatial resolution of 0.25°×0.25°
The AO (Marosz, 2010), NAO (Hurrell, 2005) and AD (Wu et al., 2006) indices all indicate variations in atmospheric pressure. The AO index is an indicator of the MSLP variability for the Northern Hemisphere above 20°N; when the Arctic pressure is low, the AO is positive. The NAO refers to the reverse change relationship of pressure between the Azores High and the Icelandic Low. When a strong subtropical high is located over the Azores and a strong low-pressure system is centred over Iceland, the NAO is positive; when weaker high-pressure and low-pressure systems are found over the same locations, the NAO is negative. We select AO and NAO data offered by the NOAA’s Climate Prediction Center from 2011 to 2018 to analyse the relationship between these atmospheric circulations and the sea ice export. The AD index corresponds to the second leading mode of empirical orthogonal function analysis based on the MSLP anomaly north of 70°N (Wu et al., 2006). We calculate the AD index from 2011 to 2018 using MSLP data.
Except for the SIC, the sea ice velocity and sea ice thickness are downloaded as Network Common Data Format (NetCDF). First, we convert the NetCDF data into a TIF format to keep the projection and spatial resolution consistent with the SIC through reprojection and resampling. According to DeRepentigny et al. (2020), we presume that a value of 251 is 100% sea ice, and convert the gray values to a range of 0%–100% by dividing by 251.
The x-axis of the downloaded sea ice velocity coincides with the 45°E meridian, and the y-axis coincides with the 135°E meridian. For calculating sea ice volume exports, the x-axis of the sea ice velocity is converted to point towards the east, and the y-axis is converted to point towards the north.
A negative volume export represents a sea ice volume loss from the Arctic Ocean (Ricker et al., 2018). The sea ice thickness, velocity and concentration data are imported into Eq. (1) and Eq. (2) to calculate the sea ice volume export.
| $$ Q_x = \sum _1^n H_xvC_xL , $$ | (1) |
| $$ Q_y = \sum_1^nH_yuC_yL , $$ | (2) |
| $$ {Q=Q}_{x} + {Q}_{y} , $$ | (3) |
where
The expected uncertainty in the sea ice volume export is calculated as follows:
| $$ {\text{δ} }_{{Q}_{x}} = {{L}}_{{x}}\sqrt{{({{H}}_{{x}}{\text{δ}}_{{v}}{{C}}_{{x}}{)}}^{{2}}+{{(}{\text{δ}}_{{{H}}_{{x}}}{v}{{C}}_{{x}}{)}}^{{2}}+{{(}{{{H}}_{{x}}{v\text{δ}}}_{{{C}}_{{x}}}{)}}^{{2}}} , $$ | (4) |
| $$ {\text{δ} }_{{Q}_{y}} = L_y\sqrt{{(H_y{\text{δ}}_uC_y)}^2+{({{\text{δ}}}_{H_y}{u}C_y)}^2+(H_yu\text{δ}_{C_y})^{{2}}} , $$ | (5) |
where
The wind stress will drive the movement of the seawater, leading to ET and Ekman pumping. ET, a horizontal movement of seawater, is caused by wind stress. Ekman pumping, a vertical movement of seawater, is caused by wind stress curl (Wang et al., 2013; Dang et al., 2022).
| $$ \tau_x=\rho _{\rm{a}}C_{\rm{d}}(W_x^2+ W_y^2)^{1/2}W_x , $$ | (6) |
| $$ \tau_y=\rho_{\rm{a}}C_{\rm{d}}(W_x^2+W_y^2)^{1/2}W_y , $$ | (7) |
where
| $$ {{Q}{Q}}_{{x}}=\tau_x/(\rho_{\rm{w}} f) , $$ | (8) |
| $$ QQ_{{y}}=-\tau_y/(\rho_{\rm{w}}f) , $$ | (9) |
where
| $$ {f}=2\varOmega \sin\theta , $$ | (10) |
where
According to Halpern (2002), ET can be calculated as follows:
| $$ {\rm{ET}}=\tau/(\rho_{\rm{w}}f) , $$ | (11) |
where
As Geyer et al. (2009) do, we use Eq. (11) to calculate the ET in ice-covered oceans.
To calculate the average sea ice thickness, velocity and concentration at zonal gate and meridional gate, we interpolate sea ice thickness, velocity and concentration data onto the zonal gate and meridional gate with a spatial resolution of 25 km×25 km.
The maximum value of the average sea ice volume export is concentrated from February to May in a given year, i.e., winter and spring, while the minimum value is concentrated from July to September, i.e., in summer (Table 1). In March 2017, the exported ice volume reached a maximum of −643.99 km3. The minimum was 26.84 km3 in February 2011, indicating that the sea ice flowed into the Arctic Ocean from the Fram Strait. The export flux at the meridional gate (accounting for 3% of the Q) is much smaller than that at the zonal gate (accounting for 97% of the Q) (Fig. 2a). Therefore, sea ice flows out to the North Atlantic from the Arctic Ocean, mainly through the zonal gate of the Fram Strait. According to the multi-year monthly average results (Fig. 2b), the exported ice volume first decreases in January, increases in February, continues to decrease after reaching its maximum in March, reaches its minimum in July, and then begins to increase. There is a small decline in August and a rapid increase in September that continues into December. The sea ice volume export decreases by approximately 100 km3 from December to January (Fig. 2b). This is due to a decrease in sea ice velocity (zonal gate: decrease 0.03 m/s; meridional gate: decrease 0.02 m/s) and sea ice thickness (zonal gate: decrease 0.11 m) in January.
| 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | |
| Jan. | −227.53 | −122.42 | −80.35 | −41.85 | −194.73 | −215.87 | −611.79 | −252.26 |
| Feb. | 26.84 | −343.20 | −129.60 | −149.29 | −571.36 | −259.18 | −229.89 | −24.00 |
| Mar. | −522.62 | −252.66 | −276.96 | −364.08 | −528.74 | −268.73 | −643.99 | −161.75 |
| Apr. | −216.12 | −354.94 | −347.20 | −565.98 | −385.36 | −297.74 | −416.06 | −228.46 |
| May | −334.43 | −362.59 | −210.71 | −359.62 | −268.48 | −270.51 | −388.01 | −66.13 |
| Jun. | −242.20 | −218.45 | −112.01 | −158.85 | −341.68 | −236.83 | −175.53 | −235.36 |
| Jul. | −81.29 | −137.02 | −82.21 | −82.65 | −145.51 | −68.31 | −27.57 | −5.64 |
| Aug. | −63.81 | −127.32 | −78.40 | −273.41 | −0.53 | −239.83 | −104.25 | −5.28 |
| Sept. | −40.05 | −89.80 | 2.74 | −240.72 | −127.48 | −178.92 | 9.42 | −139.09 |
| Oct. | −169.52 | −163.13 | −135.72 | −251.95 | −87.66 | −173.86 | −156.18 | −194.21 |
| Nov. | −184.85 | −117.19 | −295.88 | −249.67 | −258.80 | −189.79 | −170.10 | −306.79 |
| Dec. | −295.35 | −165.31 | −189.71 | −506.32 | −355.86 | −284.26 | −262.04 | −504.89 |
| Note: The absolute of underlined font indicates the maximum in a year, and the absolute of bold font indicates the minimum in a year. | ||||||||
DownLoad:
CSV
In summer, the sea ice volume export is generally lower than −100 km3. Only in 2012, 2014 and 2016 did the sea ice export flux exceed −100 km3 (Fig. 3). Two extreme cyclones occurred in August 2012 and 2016 (Simmonds and Rudeva, 2012; Yamagami et al., 2017), namely C12 and C16. The C12 originated in Siberia on August 3 and disappeared in the Canadian Archipelago on August 14, 2012, with a minimum pressure of 964.1 hPa. The C16 originated with a merging of two cyclones from northeast Siberia and the Barents Sea on August 4, 2016. Then, the C16 further merged several cyclones. Finally, the C16 disappeared in the Canadian Archipelago on September 16. The C16 showed a minimum pressure of 972.3 hPa. Great cyclones introduced strong winds and increased the sea ice velocity, prompting an increase in sea ice transport. How great cyclones affect the pressure gradient across both sides of the Fram Strait needs further study in the future. In 2014 and 2016, the winter sea ice volume export increased significantly, especially in 2016, when the flux increased by approximately 200 km3 from autumn to winter (Fig. 3). The sea ice volume export reached a maximum in winter 2016 and a minimum in summer 2017 (Fig. 3). Uncertainties in sea ice export caused by the uncertainty of sea ice thickness, velocity and concentration are generally within ±50 km3.
The annual sea ice volume export increases from 2011 to 2018, with a change rate of −4.25 km 3/a (where a denotes year). The average spring sea ice volume export is the highest (−283.05 km3), and the average summer sea ice volume export is the lowest (−96.96 km3). The spring sea ice volume exports have a change rate of 5.18 km3/a, and the winter sea ice volume exports have a change rate of −16.02 km3/a. If winter sea ice volume export changes according to this change rate, the sea ice volume exports in winter will exceed that in spring.
To verify the reliability of the sea ice volume export results in this study, we compare the results of this paper with the results of Min et al. (2019) (hereafter referred to as M19). Min et al. (2019) studied sea ice volume export through the gate A (zonal gate: at 82°N between 12°W and 20°E; meridional gate: at 20°E between 80.5°N and 82°N) from 2011 to 2016. Therefore, we calculate the sea ice volume export through the gate A, and the results from 2011 to 2016 are compared. Large difference between our monthly average ice volume exports and M19 estimates appears in July, our results are 43 km3 lower than M19 estimates (Fig. 4a). However, our monthly results agree well with the M19 estimates within one standard deviation. The differences in spring, autumn and winter are within 5 km3; the main difference is in summer, with a difference of approximately 15 km3 (Fig. 4b). There are differences in the sea ice thickness, velocity and SIC data sets with different accuracies. The accuracies of sea ice thickness, velocity and SIC during the cold season are much higher than those in summer. Therefore, the differences between different exported sea ice volume estimates in the cold season are small. There are some differences in summer sea ice volume export. The consistent results indicate that the CMEMS reanalysis data can be applied to the sea ice volume export estimation, and the result is relatively reliable.
To compare with more previous studies, we calculate the sea ice volume export through the gate B (~79°N gate). The annual average sea ice volume export through the gate A from 2011 to 2018 is 207 km3 lower than M19 estimates from 2011 to 2016. The annual average sea ice volume export through the gate B from 2011 to 2018 is 53 km3 lower than M19 estimates from 2011 to 2016 (Table 2). Both of them indicate sea ice volume export decreases during 2017–2018. Our average annual export flux through the gate B during 2011–2016 is very close to the M19 estimates, with a difference of only 22 km3 (Table 2). Compared with the average annual sea ice volume export from 1991 to 1995 (Kwok and Rothlock, 1999), the average annual export flux (1991–1998) estimated by Kwok et al. (2004) decreased by 288 km3, indicating a rapid decrease in export flux during 1996–1998 (Table 2). Compared with the average annual sea ice volume export from 1991 to 1998, the average annual export flux from 2011 to 2018 decreased by nearly 1 000 km3 (Table 2). The accelerated melting of sea ice in the recent 20 years contributes to the decrease in sea ice volume export.
| This study | Min et al. (2019) | Kwok and Rothrock (1999) | Kwok et al. (2004) | ||
| Period | 2011–2018 | 2011–2016 | 2011–2016 | 1991–1995 | 1991–1998 |
| Average volume flux/km3 | −2 096/−1 299 | −2 226/−1 374 | −2 303/−1 352 | −2 506 | −2 218 |
| Note: The underlined font indicates the sea ice volume export through the gate A, and the bold font indicates the sea ice volume export through the gate B (~79°N gate). The listed volume flux represents the average volume flux during the corresponding study period. | |||||
DownLoad:
CSV
To obtain the contributions of the sea ice thickness, sea ice velocity and SIC to the sea ice volume export, we standardize these three variables to between −2 and 2, obtain the results of comparisons among the monthly sea ice thickness, sea ice velocity and SIC changes from 2011 to 2018, and calculate their correlations with the total sea ice volume export (Fig. 5). The results show the significant positive correlation between the total sea ice volume export and the sea ice thickness, concentration and velocity (at the 0.01 significance level). The sea ice volume export is most related to the ice velocity with the largest correlation coefficient of 0.68, followed by the SIC (0.50), while the correlation coefficient between the sea ice thickness and sea ice volume export is the smallest (0.34).
To further measure the influences of these three sea ice parameters on the exported sea ice volume, we calculate the relative standard deviation (RSD=SD/Mean) (Ricker et al., 2018; Min et al., 2019).
Variables that have a larger RSD contribute greatly to the variation in the sea ice volume export. The ice velocity has the largest RSD of 0.49 with a mean velocity of 0.08 m/s (Table 3), which indicates that the ice velocity has the greatest impact on the volume export variation. The mean SIC is 60.50%, with an RSD of 0.12. The mean sea ice thickness is 1.20 m with an RSD of 0.10 (Table 3), which indicates that the thickness contributes the smallest impact on the volume export variation.
| Ice thickness | Ice velocity | Ice concentration | |
| Mean | 1.20 m | 0.08 m/s | 60.50% |
| RSD | 0.10 | 0.49 | 0.12 |
DownLoad:
CSV
After analysing the contributions of the above three variables to the sea ice volume export variation, we further analyse the respective contributions of these three sea ice variables to the seasonal variation in the sea ice volume export. We calculate the relative frequency (RF) of the seasonal sea ice thickness, sea ice velocity and SIC as follows:
| $$ {\rm{RF}}=n/N_{{\rm{grids}}} , $$ | (12) |
where n represents the number of pixels accounted for in different value ranges and Ngrids is the sum of the pixels.
From the RF distribution of the sea ice thickness at the zonal gate (Fig. 6a), we can see that the average sea ice thicknesses among the four seasons are more than 1 m, accounting for 79.51% of the total sea ice. Thin ice can be more observed in autumn (sea ice with a thickness of less than 1 m is called thin ice in this study), while thick ice (greater than 1 m) can be more observed in spring, and the maximum thickness is more than 4 m. Sea ice with a thickness of more than 2 m is distributed the most in spring, accounting for 61.11% of the spring sea ice. The thick ice benefits the largest sea ice volume export in spring. The ice thickness at the meridional gate is far less than that at the zonal gate, and the maximum is no more than 2.5 m (Fig. 6b). In winter, extremely thin ice is more abundant (ice thickness is less than 0.5 m), accounting for 76.56% of all sea ice. In contrast to the zonal gate, at which the thick ice bin has a high RF, the extremely thin ice bin at the meridional gate has high RF values in summer, autumn and winter (Fig. 6b).
The sea ice velocity at the zonal gate is generally high; more sea ice with a high drift speed can be observed in autumn, and sea ice with speeds higher than 0.06 m/s accounts for 76.85% of the total autumn sea ice (Fig. 7a). More sea ice with a low drift speed can be observed in summer, the speed bin ranging from 0.02 cm/s to 0.04 cm/s has a rather high RF, and the highest speed is not more than 0.10 m/s. Except for summer, the ice velocity in the other three seasons is mainly greater than or equal to 0.06 m/s, accounting for 69.75% of all sea ice (Fig. 7a). The sea ice velocity at the meridional gate is concentrated below 0.04 m/s, accounting for 59.77% of sea ice (Fig. 7b). In spring, the high-speed range has a high RF, but the low-speed range also has a high RF. In summer, sea ice with a speed less than 0.04 m/s has an RF as high as 95.31% (Fig. 7b). The low ice speeds at both zonal gate and meridional gate in summer contribute to low summer sea ice volume export.
The SIC at the zonal gate is high; higher concentrations of sea ice can be observed in spring, generally higher than 90%, and lower concentrations of sea ice can be observed in summer (Fig. 8a). Among the four seasons, sea ice with a concentration higher than 80% is predominant and accounts for 69.94% of the total sea ice. In spring, autumn and winter, the highest RFs appear in the highest concentration bin (90%–100%). In summer, sea ice with a concentration less than 50% is still rare, and the highest RF occurs in the range of 50%–60%. The SIC at the meridional gate is mainly less than 50% in four seasons, accounting for 64.06% (Fig. 8b). Higher concentrations of sea ice can be observed in spring. The highest RFs in summer, autumn and winter appear in the low concentration bin (0%–15%), indicating that more areas are open water. The high concentrations of ice at both zonal gate and meridional gate in spring benefit the largest spring sea ice volume export.
In summary, the zonal gate exports sea ice with a high ice thickness (mainly greater than 1 m), high ice velocity (mainly greater than 0.06 m/s) and high SIC (mainly greater than 80%). The meridional gate mainly exports sea ice with a low ice thickness (less than 0.5 m), low ice velocity (less than 0.04 m/s) and low SIC (less than 50%). Hence, the sea ice volume export through the meridional gate is far less than that through the zonal gate.
After the standardization of the AO, NAO and AD indices, we analyse the relationships between the AO, NAO, and AD and the sea ice volume export. The AO has a significant positive correlation with the sea ice volume export (Fig. 9). The greater the AO is, the greater the sea ice volume that is exported through the Fram Strait. A positive AO indicates that the Arctic is controlled by low pressure, which may result in strong winds. The movement of sea ice towards the Greenland Sea and the Norwegian Sea is accelerated, promoting an increase in the exported sea ice volume. There is a weak positive correlation between the NAO and sea ice volume export. The AD, which has a significant positive correlation with sea ice volume export with the largest correlation coefficient among three climate indices (0.44), may affect sea ice transport more directly than NAO and AO (Fig. 9). AD has a close relationship with the meridional ice drift. The strong meridional wind anomalies generated by the AD enhance the transpolar drift stream, causing the movement of sea ice from the Arctic Ocean to the Greenland Sea and Barents Sea (Zhang et al., 2022). A positive AD indicates that the Canadian Archipelago pressure anomaly is positive, the Barents Sea pressure anomaly is negative, the Beaufort Gyre is weakened, and an abnormal wind blows over the Greenland Sea, leading to an accelerated movement of sea ice towards the Greenland Sea and the Norwegian Sea, and a higher sea ice export flux through the Fram Strait. When the AD index is negative, the Beaufort Gyre strengthens, causing more sea ice to stay in the western Arctic (Ikeda, 2009).
In addition to atmospheric circulation indices, oceanic circulation indices (ET) caused by wind stress may also affect variations in sea ice volume export. After the standardization of the ET index, we analyse the relationships between the ET and the sea ice volume export (Fig. 10). The annual average ET shows a downward trend (Fig. 10a). For the ET, the average ET is the smallest in summer and shows an increasing trend. The average ET is the largest in autumn, showing a decreasing trend. The average ET in winter is second only to that in autumn, showing an increasing trend. The average ET in spring shows a decreasing trend. The changes in sea ice volume export are consistent with that in ET (Fig. 10b). The ET is low in summer and is high in winter. There is a significant positive correlation between ET and sea ice volume export flux, with a correlation coefficient of 0.51 (Fig. 10b). The higher ET, the faster the horizontal movement of seawater, resulting in faster sea ice velocity, then promoting the increase of sea ice volume export.
The sea ice volume export flux reached the maximum in winter 2016, followed by winter 2014. The sea ice volume export flux in the winters of 2012, 2013 and 2017 was small (Fig. 3). The spatial distribution of ET in winter 2012, 2013, 2014, 2016 and 2017 was obtained (Fig. 11). In winter 2016, ET in the Fram Strait was high, and ET in and around the zonal gate of Fram Strait was higher than 0.20 m2/s. In winter 2014, ET in nearly half of the area of Fram Strait was higher than 0.20 m2/s, and the high value was mainly located in the west of Fram Strait. In winters 2012, 2013 and 2017, ET in the Fram Strait was low, mainly between 0.10 m2/s and 0.20 m2/s. Among these three years, the ET was lower than 0.10 m2/s in some areas of the west of Fram Strait in winter 2012 and 2013. The spatial distribution of ET also confirms that the sea ice volume export is closely related to ET.
From the correlation analysis results, it can be seen that AD, AO and ET are significantly correlated with the sea ice volume export (Figs 9 and 10), which provides potential control factors of sea ice volume export. Multiple general linear model (GLM) analysis is performed to discuss control factors of sea ice volume export. Based on Akaike Information Criterion (AIC), the multiple GLM analysis uses the backward stepwise regression method to select the best model (Zhang et al., 2019). The model with the lowest AIC is selected and the analysis of variance (ANOVA) is performed. The proportion of variances explained by the variable is calculated according to mean squares (MS), which is used to evaluate the contribution of atmospheric circulation and oceanic circulation indices to the sea ice volume export variability. The proportion of variances explained by the variable was called SS in Tao et al. (2015).
AD contributes most to the variation in sea ice volume export, which explains 53.86% of the variation in sea ice volume export (Table 4). Followed by the ET, which has an explanatory power of 38.37% for variations in sea ice volume export (Table 4). AO accounts for an additional 5.83% (Table 4). The residuals are only 1.93%, indicating that the variation in sea ice volume export through the Fram Strait can be well controlled by the AD, ET and AO. Here, Fram Strait is defined as an area controlled by AD and ET. We can further explain the jump decline of sea ice volume flux from December to January shown in Fig. 2 according to AD and ET. The AD in January is about 1/3 of that in December, and the ET decreased by 0.18 m2/s. The decrease in AD results in the weakening of the abnormal east wind. Both two factors weaken the movement of sea ice towards the Greenland Sea and the Norwegian Sea, bringing in a large decline in sea ice volume export in January.
| AD | AO | ET | Residuals | |
| MS | 392 605 | 42 498 | 279 708 | 14 075 |
| SS/% | 53.86*** | 5.83 | 38.37*** | 1.93 |
| Note: *** denotes p<0.001; MS: mean squares. | ||||
DownLoad:
CSV
Based on sea ice thickness, sea ice velocity, and SIC data, we obtain the sea ice volume export through the Fram Strait from 2011 to 2018 and analyse the contributions of the sea ice thickness, velocity and concentration to the exported sea ice volume. Finally, we discuss the influences of atmospheric circulation indices (AO, NAO, and AD) and oceanic circulation index (ET) on the sea ice volume export. We draw the following conclusions:
(1) The zonal gate of the Fram Strait contributes 97% of the total sea ice volume export, while the meridional gate only contributes 3% of the total sea ice volume export. The uncertainty of sea ice export is generally less than ±50 km3. The sea ice volume export through the Fram Strait is highest in spring and lowest in summer. The exported sea ice volume decreases in spring and increases in winter. If the winter exported sea ice volume continues to increase at the current rate, the winter exported sea ice volume will exceed spring exported sea ice volume in the future. The exported sea ice volume is consistent with the existing exported sea ice volume estimates, and the differences are generally within 5 km3. The obtained sea ice volume exports are reliable, and CMEMS can be applied to the calculation of exported sea ice volume. Compared with sea ice volume export in the 1990s, the exported sea ice volume in the 2010s decreased by nearly 1 000 km3.
(2) Among three input variables (sea ice thickness, sea ice velocity and SIC), the sea ice velocity has the greatest influence on the sea ice volume export. The sea ice passing through the zonal gate has three characteristics, i.e., the sea ice thickness is mainly greater than 1 m, the ice velocity is mainly faster than 0.06 m/s, and the SIC is mainly higher than 80%. However, the sea ice passing through the meridional gate with three contrary characteristics, i.e., sea ice thickness is mainly less than 0.5 m, the sea ice velocity is mainly less than 0.04 m/s and SIC is mainly less than 50%. The different distribution characteristics of sea ice between the zonal gate and meridional gate lead to the sea ice outflow through the meridional gate being far less than that through the zonal gate.
(3) The sea ice volume export is most closely related to the AD among three atmospheric indices (AD, AO and NAO). The stronger the AD is, the more sea ice is exported out of the Fram Strait into the Norwegian Sea and the Greenland Sea. The AO has a significant positive correlation with the sea ice volume export through the Fram Strait, but the correlation coefficient is lower than that of the AD. The stronger the AO is, the more sea ice is exported through the Fram Strait. In addition to the atmospheric circulation indices AD and AO, the wind-driven oceanic circulation index ET plays an important role in variation in sea ice volume export. The greater the ET, the more intense the horizontal movement of seawater and the greater the exported sea ice volume. The AD and ET explain 53.86% and 38.37% of the variation in exported sea ice volume through the Fram Strait, respectively.
Acknowledgements: We thank the National Snow and Ice Data Center for SIC data, and Copernicus Marine Environment Monitoring Service for ice thickness and ice velocity data, and European Centre for Medium-Range Weather Forecasts for mean sea level pressure, u10 and v10 data, and NOAA for AO, NAO index data.|
Acciarri A, Bisci C, Cantalamessa G, et al. 2016. Anthropogenic influence on recent evolution of shorelines between the Conero Mt. and the Tronto R. mouth (southern Marche, Central Italy). CATENA, 147: 545–555. doi: 10.1016/j.catena.2016.08.018
|
|
Besset M, Anthony E J, Bouchette F. 2019. Multi-decadal variations in delta shorelines and their relationship to river sediment supply: an assessment and review. Earth-Science Reviews, 193: 199–219. doi: 10.1016/j.earscirev.2019.04.018
|
|
Chen Xiaowen, Liu Xia, Zhang Wei. 2011. Shore reclamation in Zhujiang River Esturay and its impact analysis. Journal of Hohai University (Natural Sciences) (in Chinese), 39(1): 39–43
|
|
Chu Zhongxin, Yang Xuhui, Feng Xiuli, et al. 2013. Temporal and spatial changes in coastline movement of the Yangtze Delta during 1974–2010. Journal of Asian Earth Sciences, 66: 166–174. doi: 10.1016/j.jseaes.2013.01.002
|
|
Cooper J A G, Pilkey O H. 2004. Sea-level rise and shoreline retreat: time to abandon the Bruun Rule. Global and Planetary Change, 43(3–4): 157–171
|
|
Cowart L, Walsh J P, Corbett D R. 2010. Analyzing estuarine shoreline change: a case study of cedar Island, North Carolina. Journal of Coastal Research, 26(5): 817–830
|
|
Cui Buli, Li Xiaoyan. 2011. Coastline change of the Yellow River Estuary and its response to the sediment and runoff (1976–2005). Geomorphology, 127(1–2): 32–40
|
|
Cunliffe A M, Tanski G, Radosavljevic B, et al. 2019. Rapid retreat of permafrost coastline observed with aerial drone photogrammetry. The Cryosphere, 13(5): 1513–1528. doi: 10.5194/tc-13-1513-2019
|
|
Dai Zhijun. 2021. Changjiang Riverine and Estuarine Hydro-morphodynamic Processes: in the Context of Anthropocene Era. Singapore: Springer
|
|
Dai Zhijun, Liu J T. 2013. Impacts of large dams on downstream fluvial sedimentation: an example of the Three Gorges Dam (TGD) on the Changjiang (Yangtze River). Journal of Hydrology, 480: 10–18. doi: 10.1016/j.jhydrol.2012.12.003
|
|
Dai Zhijun, Liu J T, Wei Wen, et al. 2014. Detection of the Three Gorges Dam influence on the Changjiang (Yangtze River) submerged delta. Scientific Reports, 4: 6600. doi: 10.1038/srep06600
|
|
Dai Shibao, Lu Xixi. 2014. Sediment load change in the Yangtze River (Changjiang): a review. Geomorphology, 215: 60–73. doi: 10.1016/j.geomorph.2013.05.027
|
|
Dai Zhijun, Mei Xuefei, Darby S E, et al. 2018. Fluvial sediment transfer in the Changjiang (Yangtze) river-estuary depositional system. Journal of Hydrology, 566: 719–734. doi: 10.1016/j.jhydrol.2018.09.019
|
|
Dai Shibao, Yang Shilun, Cai Aimin. 2008. Impacts of dams on the sediment flux of the Zhujiang River, southern China. CATENA, 76(1): 36–43. doi: 10.1016/j.catena.2008.08.004
|
|
Day J W, Giosan L. 2008. Survive or subside?. Nature Geoscience, 1(3): 156–157
|
|
Dean R G, Houston J R. 2016. Determining shoreline response to sea level rise. Coastal Engineering, 114: 1–8. doi: 10.1016/j.coastaleng.2016.03.009
|
|
Fanos A M. 1995. The impact of human activities on the erosion and accretion of the Nile Delta coast. Journal of Coastal Research, 11(3): 821–833
|
|
Fearnley S M, Miner M D, Kulp M, et al. 2009. Hurricane impact and recovery shoreline change analysis of the Chandeleur Islands, Louisiana, USA: 1855 to 2005. Geo-Marine Letters, 29(6): 455–466. doi: 10.1007/s00367-009-0155-5
|
|
Ford M. 2013. Shoreline changes interpreted from multi-temporal aerial photographs and high resolution satellite images: Wotje Atoll, Marshall Islands. Remote Sensing of Environment, 135: 130–140. doi: 10.1016/j.rse.2013.03.027
|
|
Frihy O E, Debes E A, El Sayed W R. 2003. Processes reshaping the Nile delta promontories of Egypt: pre- and post-protection. Geomorphology, 53(3–4): 263–279
|
|
Gellis A C, Noe G B. 2013. Sediment source analysis in the Linganore Creek watershed, Maryland, USA, using the sediment fingerprinting approach: 2008 to 2010. Journal of Soils and Sediments, 13(10): 1735–1753. doi: 10.1007/s11368-013-0771-6
|
|
Gong Wenping, Shen Jian. 2011. The response of salt intrusion to changes in river discharge and tidal mixing during the dry season in the Modaomen Estuary, China. Continental Shelf Research, 31(7–8): 769–788
|
|
Grill G, Lehner B, Thieme M, et al. 2019. Mapping the world’s free-flowing rivers. Nature, 569(7755): 215–221. doi: 10.1038/s41586-019-1111-9
|
|
Han Zhiyuan, Tian Xiangping, Ou Suying. 2010. Impacts of Large-scale human activities on riverbed morphology and tidal dynamics at modaomen estuary. Scientia Geographica Sinica (in Chinese), 30(4): 582–587
|
|
Hansen J E, Elias E, List J H, et al. 2013. Tidally influenced alongshore circulation at an inlet-adjacent shoreline. Continental Shelf Research, 56: 26–38. doi: 10.1016/j.csr.2013.01.017
|
|
Healy M G, Hickey K R. 2002. Historic land reclamation in the intertidal wetlands of the Shannon Estuary, western Ireland. Journal of Coastal Research, 36(10036): 365–373
|
|
Hillel D. 2013. Introduction to Soil Physics. New York: Academic Press
|
|
Himmelstoss E A, Henderson R E, Kratzmann M G, et al. 2018. Digital shoreline analysis system (DSAS) version 5.0 user guide. Reston: U. S. Geological Survey
|
|
Jia Liangwen, Pan Shunqi, Wu Chaoyu. 2013. Effects of the anthropogenic activities on the morphological evolution of the Modaomen Estuary, Zhujiang River Delta, China. China Ocean Engineering, 27(6): 795–808. doi: 10.1007/s13344-013-0065-1
|
|
Jones B M, Grosse G, Arp C D, et al. 2011. Modern thermokarst lake dynamics in the continuous permafrost zone, northern Seward Peninsula, Alaska. Journal of Geophysical Research: Biogeosciences, 116(G2): G00M03
|
|
Kong Dongxian, Miao Chiyuan, Borthwick A G L, et al. 2015. Evolution of the Yellow River Delta and its relationship with runoff and sediment load from 1983 to 2011. Journal of Hydrology, 520: 157–167. doi: 10.1016/j.jhydrol.2014.09.038
|
|
Lehner B, Liermann C R, Revenga C, et al. 2011. High-resolution mapping of the world’s reservoirs and dams for sustainable river-flow management. Frontiers in Ecology and the Environment, 9(9): 494–502. doi: 10.1890/100125
|
|
Liu Feng, Hu Shuai, Guo Xiaojuan, et al. 2018. Recent changes in the sediment regime of the Zhujiang River (South China): causes and implications for the Zhujiang River Delta. Hydrological Processes, 32(12): 1771–1785. doi: 10.1002/hyp.11513
|
|
Liu Feng, Tian Xiangping, Han Zhiyuan, et al. 2011. Analysis of river channel evolution of Modaomen channel of Xijiang River in past forty years. Journal of Sediment Research (in Chinese), (1): 45–50
|
|
Liu Feng, Yuan Lirong, Yang Qingshu, et al. 2014. Hydrological responses to the combined influence of diverse human activities in the Zhujiang River Delta, China. CATENA, 113: 41–55. doi: 10.1016/j.catena.2013.09.003
|
|
Loveland T R, Dwyer J L. 2012. Landsat: building a strong future. Remote Sensing of Environment, 122: 22–29. doi: 10.1016/j.rse.2011.09.022
|
|
Luo Xiangxin, Yang Shilun, Zhang Jing. 2012. The impact of the Three Gorges Dam on the downstream distribution and texture of sediments along the middle and lower Yangtze River (Changjiang) and its estuary, and subsequent sediment dispersal in the East China Sea. Geomorphology, 179: 126–140. doi: 10.1016/j.geomorph.2012.05.034
|
|
Luo Xianlin, Zeng E Y, Ji Rongyao, et al. 2007. Effects of in-channel sand excavation on the hydrology of the Zhujiang River Delta, China. Journal of Hydrology, 343(3–4): 230–239
|
|
Matin N, Hasan G M J. 2021. A quantitative analysis of shoreline changes along the coast of Bangladesh using remote sensing and GIS techniques. CATENA, 201: 105185. doi: 10.1016/j.catena.2021.105185
|
|
McFeeters S K. 1996. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. International Journal of Remote Sensing, 17(7): 1425–1432. doi: 10.1080/01431169608948714
|
|
Meade R H. 1982. Sources, sinks, and storage of river sediment in the Atlantic drainage of the United States. The Journal of Geology, 90(3): 235–252. doi: 10.1086/628677
|
|
Mei Xuefei, Dai Zhijun, Du Jinzhou, et al. 2015. Linkage between Three Gorges Dam impacts and the dramatic recessions in China’s largest freshwater lake, Poyang Lake. Scientific Reports, 5: 18197. doi: 10.1038/srep18197
|
|
Nilsson C, Reidy C A, Dynesius M, et al. 2005. Fragmentation and flow regulation of the world’s large river systems. Science, 308(5720): 405–408. doi: 10.1126/science.1107887
|
|
Pritchard D W. 1967. What is an estuary: physical viewpoint. American Association for the Advancement of Science, 83: 3–5
|
|
Roccati A, Faccini F, Luino F, et al. 2019. Morphological changes and human impact in the Entella River floodplain (Northern Italy) from the 17th century. CATENA, 182: 104122. doi: 10.1016/j.catena.2019.104122
|
|
San-Nami T, Uda T, Onaka S. 2013. Long-term shoreline recession on eastern Bali Coast caused by riverbed mining. In: Proceedings of the 7th International Conference on Asian and Pacific Coasts. Bali: Hasanuddin University, 275–282
|
|
Stanley D J, Warne A G. 1993. Nile delta: recent geological evolution and human impact. Science, 260(5108): 628–634. doi: 10.1126/science.260.5108.628
|
|
Thieler E R, Himmelstoss E A, Zichichi J L, et al. 2009. The digital shoreline analysis system (DSAS) version 4.0-An ArcGIS extension for calculating shoreline change. Reston: U. S. Geological Survey
|
|
van der Wal D, Pye K, Neal A. 2002. Long-term morphological change in the Ribble Estuary, northwest England. Marine Geology, 189(3–4): 249–266
|
|
Wang Jie, Dai Zhijun, Mei Xuefei, et al. 2020. Tropical cyclones significantly alleviate mega—deltaic erosion induced by high riverine flow. Geophysical Research Letters, 47(19): e2020GL089065
|
|
Wei Xing, Cai Shuqun, Zhan Weikang. 2021. Impact of anthropogenic activities on morphological and deposition flux changes in the Zhujiang River Estuary, China. Scientific Reports, 11(1): 16643. doi: 10.1038/s41598-021-96183-0
|
|
Wu Ziyin, Milliman J D, Zhao Dineng, et al. 2018. Geomorphologic changes in the lower Zhujiang River Delta, 1850–2015, largely due to human activity. Geomorphology, 314: 42–54. doi: 10.1016/j.geomorph.2018.05.001
|
|
Wu Ziyin, Saito Y, Zhao Dineng, et al. 2016a. Impact of human activities on subaqueous topographic change in Lingding Bay of the Zhujiang River Estuary, China, during 1955–2013. Scientific Reports, 6: 37742. doi: 10.1038/srep37742
|
|
Wu Chuangshou, Yang Shilun, Huang Shichang, et al. 2016b. Delta changes in the Zhujiang River Estuary and its response to human activities (1954–2008). Quaternary International, 392: 147–154. doi: 10.1016/j.quaint.2015.04.009
|
|
Yang Shilun, Belkin I M, Belkina A I, et al. 2003. Delta response to decline in sediment supply from the Yangtze River: evidence of the recent four decades and expectations for the next half-century. Estuarine, Coastal and Shelf Science, 57(4): 689–699
|
|
Yang Liuzhu, Liu Feng, Gong Wenping, et al. 2019. Morphological response of Lingding Bay in the Zhujiang River Estuary to human intervention in recent decades. Ocean & Coastal Management, 176: 1–10
|
|
Yang Shilun, Milliman J D, Li Peng, et al. 2011. 50, 000 dams later: erosion of the Yangtze River and its delta. Global and Planetary Change, 75(1–2): 14–20
|
|
Yang Zuosheng, Wang Houjie, Saito Y, et al. 2006. Dam impacts on the Changjiang (Yangtze) River sediment discharge to the sea: the past 55 years and after the Three Gorges Dam. Water Resources Research, 42(4): W04407
|
|
Yang Shilun, Zhang Jianbo, Xu Xiaojun. 2007. Influence of the Three Gorges Dam on downstream delivery of sediment and its environmental implications, Yangtze River. Geophysical Research Letters, 34(10): L10401. doi: 10.1029/2007GL029472
|
|
Zhang Shurong, Lu Xixi. 2009. Hydrological responses to precipitation variation and diverse human activities in a mountainous tributary of the lower Xijiang, China. CATENA, 77(2): 130–142. doi: 10.1016/j.catena.2008.09.001
|
|
Zhang Qiang, Xu Chongyu, Chen Yongqin David, et al. 2009. Abrupt behaviors of the streamflow of the Zhujiang River basin and implications for hydrological alterations across the Zhujiang River Delta, China. Journal of Hydrology, 377(3–4): 274–283
|
|
Zhang Wei, Xu Yang, Hoitink A J F, et al. 2015. Morphological change in the Zhujiang River Delta, China. Marine Geology, 363: 202–219. doi: 10.1016/j.margeo.2015.02.012
|
|
Zhu Meisha, Sun Tao, Shao Dongdong. 2016. Impact of land reclamation on the evolution of shoreline change and nearshore vegetation distribution in Yangtze River Estuary. Wetlands, 36(S1): 11–17. doi: 10.1007/s13157-014-0610-6
|
Figures(11) / Tables(8)
Supported by:
Beijing Renhe Information Technology Co. Ltd
Haili Li, Changqing Ke, Qinghui Zhu, Xiaoyi Shen. Arctic sea ice volume export through the Fram Strait: variation and its effect factors[J]. Acta Oceanologica Sinica, 2023, 42(5): 166-178. doi: 10.1007/s13131-022-2075-4
| 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | |
| Jan. | −227.53 | −122.42 | −80.35 | −41.85 | −194.73 | −215.87 | −611.79 | −252.26 |
| Feb. | 26.84 | −343.20 | −129.60 | −149.29 | −571.36 | −259.18 | −229.89 | −24.00 |
| Mar. | −522.62 | −252.66 | −276.96 | −364.08 | −528.74 | −268.73 | −643.99 | −161.75 |
| Apr. | −216.12 | −354.94 | −347.20 | −565.98 | −385.36 | −297.74 | −416.06 | −228.46 |
| May | −334.43 | −362.59 | −210.71 | −359.62 | −268.48 | −270.51 | −388.01 | −66.13 |
| Jun. | −242.20 | −218.45 | −112.01 | −158.85 | −341.68 | −236.83 | −175.53 | −235.36 |
| Jul. | −81.29 | −137.02 | −82.21 | −82.65 | −145.51 | −68.31 | −27.57 | −5.64 |
| Aug. | −63.81 | −127.32 | −78.40 | −273.41 | −0.53 | −239.83 | −104.25 | −5.28 |
| Sept. | −40.05 | −89.80 | 2.74 | −240.72 | −127.48 | −178.92 | 9.42 | −139.09 |
| Oct. | −169.52 | −163.13 | −135.72 | −251.95 | −87.66 | −173.86 | −156.18 | −194.21 |
| Nov. | −184.85 | −117.19 | −295.88 | −249.67 | −258.80 | −189.79 | −170.10 | −306.79 |
| Dec. | −295.35 | −165.31 | −189.71 | −506.32 | −355.86 | −284.26 | −262.04 | −504.89 |
| Note: The absolute of underlined font indicates the maximum in a year, and the absolute of bold font indicates the minimum in a year. | ||||||||
DownLoad:
CSV
| This study | Min et al. (2019) | Kwok and Rothrock (1999) | Kwok et al. (2004) | ||
| Period | 2011–2018 | 2011–2016 | 2011–2016 | 1991–1995 | 1991–1998 |
| Average volume flux/km3 | −2 096/−1 299 | −2 226/−1 374 | −2 303/−1 352 | −2 506 | −2 218 |
| Note: The underlined font indicates the sea ice volume export through the gate A, and the bold font indicates the sea ice volume export through the gate B (~79°N gate). The listed volume flux represents the average volume flux during the corresponding study period. | |||||
DownLoad:
CSV
| Ice thickness | Ice velocity | Ice concentration | |
| Mean | 1.20 m | 0.08 m/s | 60.50% |
| RSD | 0.10 | 0.49 | 0.12 |
DownLoad:
CSV
| AD | AO | ET | Residuals | |
| MS | 392 605 | 42 498 | 279 708 | 14 075 |
| SS/% | 53.86*** | 5.83 | 38.37*** | 1.93 |
| Note: *** denotes p<0.001; MS: mean squares. | ||||
DownLoad:
CSV
| 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | |
| Jan. | −227.53 | −122.42 | −80.35 | −41.85 | −194.73 | −215.87 | −611.79 | −252.26 |
| Feb. | 26.84 | −343.20 | −129.60 | −149.29 | −571.36 | −259.18 | −229.89 | −24.00 |
| Mar. | −522.62 | −252.66 | −276.96 | −364.08 | −528.74 | −268.73 | −643.99 | −161.75 |
| Apr. | −216.12 | −354.94 | −347.20 | −565.98 | −385.36 | −297.74 | −416.06 | −228.46 |
| May | −334.43 | −362.59 | −210.71 | −359.62 | −268.48 | −270.51 | −388.01 | −66.13 |
| Jun. | −242.20 | −218.45 | −112.01 | −158.85 | −341.68 | −236.83 | −175.53 | −235.36 |
| Jul. | −81.29 | −137.02 | −82.21 | −82.65 | −145.51 | −68.31 | −27.57 | −5.64 |
| Aug. | −63.81 | −127.32 | −78.40 | −273.41 | −0.53 | −239.83 | −104.25 | −5.28 |
| Sept. | −40.05 | −89.80 | 2.74 | −240.72 | −127.48 | −178.92 | 9.42 | −139.09 |
| Oct. | −169.52 | −163.13 | −135.72 | −251.95 | −87.66 | −173.86 | −156.18 | −194.21 |
| Nov. | −184.85 | −117.19 | −295.88 | −249.67 | −258.80 | −189.79 | −170.10 | −306.79 |
| Dec. | −295.35 | −165.31 | −189.71 | −506.32 | −355.86 | −284.26 | −262.04 | −504.89 |
| Note: The absolute of underlined font indicates the maximum in a year, and the absolute of bold font indicates the minimum in a year. | ||||||||
| This study | Min et al. (2019) | Kwok and Rothrock (1999) | Kwok et al. (2004) | ||
| Period | 2011–2018 | 2011–2016 | 2011–2016 | 1991–1995 | 1991–1998 |
| Average volume flux/km3 | −2 096/−1 299 | −2 226/−1 374 | −2 303/−1 352 | −2 506 | −2 218 |
| Note: The underlined font indicates the sea ice volume export through the gate A, and the bold font indicates the sea ice volume export through the gate B (~79°N gate). The listed volume flux represents the average volume flux during the corresponding study period. | |||||
| Ice thickness | Ice velocity | Ice concentration | |
| Mean | 1.20 m | 0.08 m/s | 60.50% |
| RSD | 0.10 | 0.49 | 0.12 |
| AD | AO | ET | Residuals | |
| MS | 392 605 | 42 498 | 279 708 | 14 075 |
| SS/% | 53.86*** | 5.83 | 38.37*** | 1.93 |
| Note: *** denotes p<0.001; MS: mean squares. | ||||
DownLoad:
