A Novel Data Fusion Technique for Snow Cover Retrieval

Gregorio, Ludovica De; Callegari, Mattia; Marín, Carlo; Zebisch, Marc; Bruzzone, Lorenzo; Demir, Begüm; Strasser, Ulrich; Marke, Thomas; Günther, Daniel; Nadalet, Rudi; Notarnicola, Claudia

doi:10.1109/jstars.2019.2920676

Cited by 16 publications

(8 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Support Vector Regression (SVR) is a supervised learning algorithm for regression [48,49]. SVR relies on establishing a regression function, and SVR is a statistical learning theorybased machine learning formalism.…”

Section: Machine Learning Methodsmentioning

confidence: 99%

Snow Depth Fusion Based on Machine Learning Methods for the Northern Hemisphere

Che

Dai

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

In this study, a machine learning algorithm was introduced to fuse gridded snow depth datasets. The input variables of the machine learning method included geolocation (latitude and longitude), topographic data (elevation), gridded snow depth datasets and in situ observations. A total of 29,565 in situ observations were used to train and optimize the machine learning algorithm. A total of five gridded snow depth datasets—Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR-E) snow depth, Global Snow Monitoring for Climate Research (GlobSnow) snow depth, Long time series of daily snow depth over the Northern Hemisphere (NHSD) snow depth, ERA-Interim snow depth and Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2) snow depth—were used as input variables. The first three snow depth datasets are retrieved from passive microwave brightness temperature or assimilation with in situ observations, while the last two are snow depth datasets obtained from meteorological reanalysis data with a land surface model and data assimilation system. Then, three machine learning methods, i.e., Artificial Neural Networks (ANN), Support Vector Regression (SVR), and Random Forest Regression (RFR), were used to produce a fused snow depth dataset from 2002 to 2004. The RFR model performed best and was thus used to produce a new snow depth product from the fusion of the five snow depth datasets and auxiliary data over the Northern Hemisphere from 2002 to 2011. The fused snow-depth product was verified at five well-known snow observation sites. The R2 of Sodankylä, Old Aspen, and Reynolds Mountains East were 0.88, 0.69, and 0.63, respectively. At the Swamp Angel Study Plot and Weissfluhjoch observation sites, which have an average snow depth exceeding 200 cm, the fused snow depth did not perform well. The spatial patterns of the average snow depth were analyzed seasonally, and the average snow depths of autumn, winter, and spring were 5.7, 25.8, and 21.5 cm, respectively. In the future, random forest regression will be used to produce a long time series of a fused snow depth dataset over the Northern Hemisphere or other specific regions.

show abstract

Section: Machine Learning Methodsmentioning

confidence: 99%

Snow Depth Fusion Based on Machine Learning Methods for the Northern Hemisphere

Che

Dai

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Machine-learning and deep-learning technology power many aspects of remote sensing: from target recognition [25][26][27][28][29] to semantic segmentation [30,31] to spatial-temporal prediction [32,33], and they are increasingly present in snow parameter estimation and retrieval [34][35][36]. Tedesco [37] constructed a neural network to retrieve the snow depth and SWE, which showed the highest accuracy compared with the other four retrieval algorithms.…”

Section: Introductionmentioning

confidence: 99%

Downscaling Snow Depth Mapping by Fusion of Microwave and Optical Remote-Sensing Data Based on Deep Learning

et al. 2021

View full text Add to dashboard Cite

Accurate high spatial resolution snow depth mapping in arid and semi-arid regions is of great importance for snow disaster assessment and hydrological modeling. However, due to the complex topography and low spatial-resolution microwave remote-sensing data, the existing snow depth datasets have large errors and uncertainty, and actual spatiotemporal heterogeneity of snow depth cannot be effectively detected. This paper proposed a deep learning approach based on downscaling snow depth retrieval by fusion of satellite remote-sensing data with multiple spatial scales and diverse characteristics. The (Fengyun-3 Microwave Radiation Imager) FY-3 MWRI data were downscaled to 500 m resolution to match Moderate-resolution Imaging Spectroradiometer (MODIS) snow cover, meteorological and geographic data. A deep neural network was constructed to capture detailed spectral and radiation signals and trained to retrieve the higher spatial resolution snow depth from the aforementioned input data and ground observation. Verified by in situ measurements, downscaled snow depth has the lowest root mean square error (RMSE) and mean absolute error (MAE) (8.16 cm, 4.73 cm respectively) among Environmental and Ecological Science Data Center for West China Snow Depth (WESTDC_SD, 9.38 cm and 5.36 cm), the Microwave Radiation Imager (MWRI) Ascend Snow Depth (MWRI_A_SD, 9.45 cm and 5.49 cm) and MWRI Descend Snow Depth (MWRI_D_SD, 10.55 cm and 6.13 cm) in the study area. Meanwhile, downscaled snow depth could provide more detailed information in spatial distribution, which has been used to analyze the decrease of retrieval accuracy by various topography factors.

show abstract

“…Satellite images permitted the description of glacier albedo [48][49][50], which is diminishing in the Alps [51] due to increasing debris and black carbon deposition. Furthermore, remote sensing highly improved snow cover mapping, thus permitting researchers to describe snow depletion and to compute more accurately the water budget of a mountain catchment [52][53][54][55][56][57][58][59][60].…”

Section: Introductionmentioning

confidence: 99%

Comparing Measured Incoming Shortwave and Longwave Radiation on a Glacier Surface with Estimated Records from Satellite and Off-Glacier Observations: A Case Study for the Forni Glacier, Italy

et al. 2020

View full text Add to dashboard Cite

The development of methods for quantifying meltwater from glaciated areas is very important for better management of water resources and because of the strong impact of current and expected climate change on the Alpine cryosphere. Radiative fluxes are the main melt-drivers, but they can generally not be derived from in situ measures because glaciers are usually located in remote areas where the number of meteorological stations is very low. For this reason, focusing, as a case study, on one of the few glaciers with a supraglacial automatic weather station (Forni Glacier), we investigated methods based on both satellite records and off-glacier surface observations to estimate incoming short- and long-wave radiation at the glacier surface (SWin and LWin). Specifically, for SWin, we considered CM SAF SARAH satellite gridded surface solar irradiance fields and data modeled by cloud transmissivity parametrized from both CM SAF COMET satellite cloud fractional cover fields and daily temperature range observed at the closest off-glacier station. We then used the latter two data sources to derive LWin too. Finally, we used the estimated SWin and LWin records to assess the errors obtained when introducing estimated rather than measured incoming radiation data to quantify glacier melting by means of an energy balance model. Our results suggest that estimated SWin and LWin records derived from satellite measures are in better agreement with in situ observations than estimated SWin and LWin records parametrized from observations performed at the closest off-glacier station. Moreover, we find that the former estimated records permit a significantly better quantification of glacier melting than the latter estimated ones.

show abstract

A Novel Data Fusion Technique for Snow Cover Retrieval

Cited by 16 publications

References 44 publications

Snow Depth Fusion Based on Machine Learning Methods for the Northern Hemisphere

Snow Depth Fusion Based on Machine Learning Methods for the Northern Hemisphere

Downscaling Snow Depth Mapping by Fusion of Microwave and Optical Remote-Sensing Data Based on Deep Learning

Comparing Measured Incoming Shortwave and Longwave Radiation on a Glacier Surface with Estimated Records from Satellite and Off-Glacier Observations: A Case Study for the Forni Glacier, Italy

Contact Info

Product

Resources

About