A deep learning approach to retrieve cold-season snow depth over Arctic sea ice from AMSR2 measurements

Li, Haili; Ke, Chang‐Qing; Zhu, Qinghui; Li, Mengmeng; Shen, Xueqin

doi:10.1016/j.rse.2021.112840

Cited by 18 publications

(10 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Li et al. (2022) and Li and Ke (2023) developed a novel method to retrieve snow depth over Arctic sea ice. This model was developed based on Advanced Microwave Scanning Radiometer 2 (AMSR2) measurements and ice mass balance buoy (IMB) data.…”

Section: Methodsmentioning

confidence: 99%

“…Based on these relationships, we expect to enhance future snow depth accuracy, clarify corresponding change patterns in snow depth, and provide some help for snow treatment via sea ice models. Here, using snow depth data sets (satellite observations) established in previous studies (Li & Ke, 2023; Li et al., 2022), the relationships between snow depth and environmental factors were investigated, and a snow depth prediction model was established. Then, snow depth data (2015–2100) were generated using a developed prediction model and CMIP6‐simulated data (environmental factors) under four CMIP6 Representative Concentration Pathway‐Shared Socioeconomic Pathway (SSP‐RCP) scenarios.…”

Section: Introductionmentioning

confidence: 99%

“…For different ice types, the decreasing trend in snow depth over first‐year ice (FYI) may be related to precipitation and ice freezing dates (Webster et al., 2014, 2018); the negative trend in snow depth in the mixed region corresponds to the decreasing trend in the proportion of MYI, and the change in sea ice type accounts for the decrease in snow depth (Lee et al., 2021). Changes in cold season snow depth in the Central Arctic are mostly influenced by air temperature and sea surface temperature (Li et al., 2022). The ice dynamics process affects the spatial variability of snow depth (J. Stroeve et al., 2020).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Ensemble Learning Model Reveals Accelerated Reductions in Snow Depth Over Arctic Sea Ice Under High‐Emission Scenarios

Li,

Ke,

Shen

et al. 2024

JGR Atmospheres

View full text Add to dashboard Cite

There are significant differences in snow depth predictions among different earth system models, and many models underestimate snow depth, restricting their application. Here, major factors influencing snow depth changes in the Coupled Model Intercomparison Project Phase 6 (CMIP6) were identified and evaluated. Based on satellite‐derived snow depth and CMIP6 data, an ensemble learning model based on multiple deep learning methods (hereafter referred to as the Multi‐DL model) was developed to predict future snow depth. According to satellite observations and two Operation IceBridge products, the Multi‐DL model yielded root mean square errors of 7.48, 6.20, and 6.17 cm. A continuous decrease in snow depth was observed from 2002 to 2100, and the rate of decrease accelerated with increasing emissions. Under the highest emission scenario, the first snow‐free year occurred in 2047, within the same decade as the first ice‐free year (2056). The predicted warm season snow depth was sensitive to sea ice velocity, sea ice concentration (siconc), precipitation, sea surface temperature (tos) and albedo, while the predicted cold season snow depth was sensitive to tos, air temperature, and siconc. The above parameters introduce some snow depth uncertainty. This method provides new ideas for predicting snow depth, and the generated snow depth records can provide data support for formulating Arctic‐related policies.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Ensemble Learning Model Reveals Accelerated Reductions in Snow Depth Over Arctic Sea Ice Under High‐Emission Scenarios

Li,

Ke,

Shen

et al. 2024

JGR Atmospheres

View full text Add to dashboard Cite

show abstract

“…Due to the orbit inclination (98.75 • in FY-3D), the study area is confined to 66 • N-88 • N. According to the National Snow and Ice Data Center (NSIDC), the Arctic includes 10 sub-regions, namely, Beaufort Sea, Chukchi Sea, East Siberian Sea, Laptev Sea, Kara Sea, Barents Sea, East Greenland Sea, Baffin Bay and Davis Strait, Canadian Archipelago, and Central Arctic (Figure 1a). [22][23][24]. Although various methods have been employed to estimate the snow depth on Arctic sea ice, there are still notable discrepancies in the results produced by different algorithms, particularly when it comes to snow on MYI.…”

Section: Study Areamentioning

confidence: 99%

“…Furthermore, as this dataset covers the entire year, it can be effectively utilized for developing inversion algorithms for other seasons. Further, Kilic et al, Liu et al, and Li et al compared it with AMSR2 data and developed a robust algorithm for retrieving snow depth on Arctic Sea ice (Ki19 and Li22) [22][23][24].…”

Section: Introductionmentioning

confidence: 99%

Retrieval of Snow Depths on Arctic Sea Ice in the Cold Season from FY-3D/MWRI Data

Yin,

He,

Sun

2024

Remote Sensing

View full text Add to dashboard Cite

Snow depth is a crucial factor in the formation of snow, and its fluctuations play a significant role in the Earth’s climate system. The existing snow depth algorithms currently lack systematic quantitative evaluation, and most of them are not suitable for direct application to Chinese satellites. Therefore, a quantitative evaluation of four existing snow depth algorithms from the Advanced Microwave Scanning Radiometer 2 (AMSR2) was conducted by comparing their estimates with the measured dataset from the Operation IceBridge project (OIB). The study found that the algorithm developed by Rostosky et al. outperforms the other three algorithms in terms of correlation. However, it is unable to accurately retrieve both high and low snow depths. On the other hand, the algorithms developed by Comiso et al. and Li et al. demonstrated strong performance in correlation and statistical characteristics. Based on these results, these two algorithms were fused to enhance the accuracy of the final algorithm. The algorithm was applied to FengYun-3D/Microwave Radiation Imager (FY-3D/MWRI) data after calibration to develop a snow depth retrieval algorithm suitable for MWRI. Validation using the 2019 OIB data indicated that the algorithm had a bias and RMSE of 1 cm and 9 cm, respectively, for first-year ice (FYI) and 3 cm and 9 cm, respectively, for multi-year ice (MYI).

show abstract

Widespread Decline of the Warm Season Snow Depth Over Arctic Sea Ice Revealed by Satellite Passive Microwave Measurements

Li,

Ke,

Zhu

et al. 2024

Intl Journal of Climatology

View full text Add to dashboard Cite

Summer snow plays an essential role in Arctic hydrology and in maintaining mass and energy balance of sea ice. However, there are great challenges in retrieving long‐term summer snow depths over Arctic sea ice. Here, we proposed a combined novel five‐variable long short‐term memory (hereafter CN5VLSTM) model based on brightness temperature data to yield warm‐season snow depth estimates. Then, year‐round snow depth estimates were obtained for the first time. The CN5VLSTM model and five additional snow depth methods were assessed during the warm season based on the ice mass balance buoy (IMB), Alfred Wegener Institute (AWI) snow buoy (AWI‐SB) and Multidisciplinary Drifting Observatory for the Study of Arctic Climate (MOSAiC) snow buoy (MOSAiC‐SB). According to the three buoy products, the accuracy of the CN5VLSTM‐derived snow depth was highest among the five snow depth estimates with RMSEs of 10.2, 16.4, and 10.1 cm, respectively. Except for in May, the Arctic snow depth showed mainly a downward trend in warm months, and a significant downward trend was found in the Central Arctic. Excluding the Barents Sea, Kara Sea and Canadian Archipelago, the average year‐round snow depth decreased in the other subregions, and a significant negative trend was observed in the East Siberian and Chukchi Seas. Snowfall was an important factor that was related to the changes in snow depth in the East Siberian and Chukchi Seas. This study can provide new insights into the evolution characteristics of summer snow depth.

show abstract

A deep learning approach to retrieve cold-season snow depth over Arctic sea ice from AMSR2 measurements

Cited by 18 publications

References 56 publications

An Ensemble Learning Model Reveals Accelerated Reductions in Snow Depth Over Arctic Sea Ice Under High‐Emission Scenarios

An Ensemble Learning Model Reveals Accelerated Reductions in Snow Depth Over Arctic Sea Ice Under High‐Emission Scenarios

Retrieval of Snow Depths on Arctic Sea Ice in the Cold Season from FY-3D/MWRI Data

Widespread Decline of the Warm Season Snow Depth Over Arctic Sea Ice Revealed by Satellite Passive Microwave Measurements

Contact Info

Product

Resources

About