Snow Coverage Mapping by Learning from Sentinel-2 Satellite Multispectral Images via Machine Learning Algorithms

Wang, Yucheng; Su, Jinya; Zhai, Xiaojun; Meng, Fanyong; Liu, Cunjia

doi:10.3390/rs14030782

“…Under the assumption that applying an NDSI of 0.4 underestimates the snow-covered area, this approach leads to an overestimate of Sentinel-2-based RSLE, which decreases the discrepancy between our webcam-based estimation and Sentinel-2 estimation further. Finally, it must be considered that satellite-based snow cover mapping is constantly improved and new methodologies based on extended spectral characteristics or deep learning have been developed, e.g., [51,52]. It would be worthwhile investigating webcam RSLE differences with such products in the future.…”

Section: Discussionmentioning

confidence: 99%

Estimating Regional Snow Line Elevation Using Public Webcam Images

Portenier

¹

,

Hasler

²

,

Wunderle

³

2022

Remote Sensing

1

0

View full text Add to dashboard Cite

Snow cover is of high relevance for the Earth’s climate system, and its variability plays a key role in alpine hydrology, ecology, and socioeconomic systems. Measurements obtained by optical satellite remote sensing are an essential source for quantifying snow cover variability from a local to global scale. However, the temporal resolution of such measurements is often affected by persistent cloud coverage, limiting the application of high resolution snow cover mapping. In this study, we derive the regional snow line elevation in an alpine catchment area using public webcams. We compare our results to the snow line information derived from the Moderate-Resolution Imaging Spectroradiometer (MODIS) and Sentinel-2 snow cover products and find our results to be in good agreement therewith. Between October 2017 and the end of June 2018, snow lines derived from webcams lie on average 55.8 m below and 33.7 m above MODIS snow lines using a normalized-difference snow index (NDSI) of 0.4 and 0.1, respectively, and are on average 53.1 m below snow lines derived from Sentinel-2. We further analyze the superior temporal resolution of webcam-based snow cover information and demonstrate its effectiveness in filling temporal gaps in satellite-based measurements caused by cloud cover. Our findings show the ability of webcam-based snow line elevation retrieval to complement and improve satellite-based measurements.

show abstract

“…The study conducted by Wang demonstrates the superior performance of both conventional machine-learning and advanced deep-learning methods over the existing rule-based Sen2Cor product for snow mapping (40). They highlighted the effectiveness of the U-Net model with four informative bands (B2, B11, B4, and B9) as inputs.…”

Section: E Snow Coverage Mapping and Fractional Snow Covermentioning

confidence: 98%

“…This 365 model was used by collecting and labeling a large dataset for snow coverage mapping, and the methodology involved exploring various input band combinations, comparing reflectance distributions, and training and comparing RF and U-Net models. The results of the study suggest that both conventional machine-learning and advanced deep-learning methods are effective for snow mapping, but the U-Net model with four informative bands as inputs outperforms other methods (40). Researchers explored the optimal selection of image pairs from MODIS and Landsat 8 OLI 370 for fractional snow cover mapping, (53) with a focus on selecting pairs with similar acquisition dates.…”

Section: E Snow Coverage Mapping and Fractional Snow Covermentioning

confidence: 99%

Advances in Image-Based Estimation of Snow Hydrology Parameters: A Systematic Literature Review

Demil,

Torabi Haghighi,

Klöve

et al. 2024

Preprint

0

View full text Add to dashboard Cite

Abstract. Accurately estimating snow hydrology parameters, including snow coverage mapping and snow depth, plays a significant role in comprehending water resource dynamics, flood forecasting, and environmental management in regions influenced by snow cover. These parameters are critical for hydrological models that simulate snowmelt and runoff, which are essential for predicting water availability and managing water resources in snow-covered areas. Traditional methods for estimating these parameters often rely on manual measurements or simplistic models, which can be inadequate for capturing the complexity of snow-related processes. In recent years, there has been a growing interest in leveraging deep learning techniques for snow hydrology parameter estimation, offering the potential to overcome these limitations. This review paper comprehensively analyzes the current state, challenges, and future directions of image-based approaches in snow hydrology parameter estimation. By harnessing the power of automated methods, particularly deep learning, these approaches demonstrate the ability to capture intricate spatial and temporal relationships present in image data. A comparative analysis between traditional and image-based methods highlights the strengths of automated approaches, including scalability and accuracy. Integration of image sensors, such as satellite imagery and crowd-sourced data, is explored as a crucial component of snow hydrology parameter estimation. Various satellite image sources, including Sentinel 1-2, Landsat, and MODIS, are discussed in terms of their suitability for snow hydrological applications. Despite the promise of image-based approaches, challenges remain, including data availability, model interpretability, and transferability. The paper identifies future research directions, emphasizing the exploration of novel deep learning architectures and uncertainty quantification techniques to address these challenges. In conclusion, this review underscores the importance of image-based approaches for advancing snow hydrology parameter estimation. By addressing challenges and maximizing potential impact, these approaches have the potential to revolutionize snow hydrological modeling and environmental management.

show abstract

“…The study conducted by Wang demonstrates the superior performance of both conventional machine-learning and advanced deep-learning methods over the existing rule-based Sen2Cor product for snow mapping (40). They highlighted the effectiveness of the U-Net model with four informative bands (B2, B11, B4, and B9) as inputs.…”

Section: E Snow Coverage Mapping and Fractional Snow Covermentioning

confidence: 98%

Advances in Image-Based Estimation of Snow Hydrology Parameters: A Systematic Literature Review

Demil,

Torabi Haghighi,

Klöve

et al. 2024

Preprint

0

View full text Add to dashboard Cite

Abstract. Accurately estimating snow hydrology parameters, including snow coverage mapping and snow depth, plays a significant role in comprehending water resource dynamics, flood forecasting, and environmental management in regions influenced by snow cover. These parameters are critical for hydrological models that simulate snowmelt and runoff, which are essential for predicting water availability and managing water resources in snow-covered areas. Traditional methods for estimating these parameters often rely on manual measurements or simplistic models, which can be inadequate for capturing the complexity of snow-related processes. In recent years, there has been a growing interest in leveraging deep learning techniques for snow hydrology parameter estimation, offering the potential to overcome these limitations. This review paper comprehensively analyzes the current state, challenges, and future directions of image-based approaches in snow hydrology parameter estimation. By harnessing the power of automated methods, particularly deep learning, these approaches demonstrate the ability to capture intricate spatial and temporal relationships present in image data. A comparative analysis between traditional and image-based methods highlights the strengths of automated approaches, including scalability and accuracy. Integration of image sensors, such as satellite imagery and crowd-sourced data, is explored as a crucial component of snow hydrology parameter estimation. Various satellite image sources, including Sentinel 1-2, Landsat, and MODIS, are discussed in terms of their suitability for snow hydrological applications. Despite the promise of image-based approaches, challenges remain, including data availability, model interpretability, and transferability. The paper identifies future research directions, emphasizing the exploration of novel deep learning architectures and uncertainty quantification techniques to address these challenges. In conclusion, this review underscores the importance of image-based approaches for advancing snow hydrology parameter estimation. By addressing challenges and maximizing potential impact, these approaches have the potential to revolutionize snow hydrological modeling and environmental management.

show abstract

Snow Coverage Mapping by Learning from Sentinel-2 Satellite Multispectral Images via Machine Learning Algorithms

Cited by 18 publications

References 28 publications

Estimating Regional Snow Line Elevation Using Public Webcam Images

Estimating Regional Snow Line Elevation Using Public Webcam Images

Advances in Image-Based Estimation of Snow Hydrology Parameters: A Systematic Literature Review

Advances in Image-Based Estimation of Snow Hydrology Parameters: A Systematic Literature Review

Contact Info

Product

Resources

About