A combined data assimilation and deep learning approach for continuous spatio-temporal SWE reconstruction from sparse ground tracks

IF 3.1 Q2 GEOSCIENCES, MULTIDISCIPLINARY Journal of Hydrology X Pub Date : 2024-10-10 DOI:10.1016/j.hydroa.2024.100190
Matteo Guidicelli , Kristoffer Aalstad , Désirée Treichler , Nadine Salzmann
{"title":"A combined data assimilation and deep learning approach for continuous spatio-temporal SWE reconstruction from sparse ground tracks","authors":"Matteo Guidicelli ,&nbsp;Kristoffer Aalstad ,&nbsp;Désirée Treichler ,&nbsp;Nadine Salzmann","doi":"10.1016/j.hydroa.2024.100190","DOIUrl":null,"url":null,"abstract":"<div><div>Our understanding of the impact of climate change on water availability and natural hazards in high-mountain regions is limited due to the spatial and temporal scarcity of ground observations of precipitation and snow. Freely available, satellite-based information about the snowpack is currently mainly limited to indirect measurements of snow-covered area or very coarse-scale snow water equivalent (SWE), but only for flat areas in lowlands without vegetation cover. Novel space-based laser altimeters, such as ICESat-2, have the potential to provide high-resolution snow depth data in worldwide mountain regions where no ground observations exist. However, these space-based laser altimeters come with spatial gaps between ground tracks, obtained without repetition at a give location. To overcome these drawbacks, here, we present a combined probabilistic data assimilation and deep learning approach to reconstruct spatio-temporal SWE from observations of snow depth along ground tracks, imitating ICESat-2 tracks in view of a potential future global application.</div><div>Our approach is based on assimilating SWE and snow cover information in a degree-day model with an iterative ensemble smoother (IES) which allows temporally reconstructing SWE along hypothetical ground tracks separated by 3 km. As input, the degree-day model uses daily precipitation and downscaled air temperature from the ERA5 reanalysis. A feedforward neural network (FNN) is then used for spatial propagation of the daily mean and standard deviation of the updated SWE ensemble members obtained from the IES. The combined IES-FNN approach provides uncertainty-aware spatio-temporally continuous estimates of SWE.</div><div>We tested our approach in the alpine Dischma valley (Switzerland) using high-resolution snow depth maps obtained from photogrammetric techniques mounted on airplanes and unmanned aerial system observations. Our results show that the IES-FNN model provides reliable estimates at a resolution of approximately 100 m. Even assimilating only one SWE observation during the year (combined with satellite-based melt-out date estimates) produces satisfying results when evaluating the IES-FNN SWE reconstructions on independent dates and smaller (<span><math><mrow><mo>&lt;</mo></mrow></math></span>4 km<sup>2</sup>) areas: mean absolute error of 86 mm (78 mm) at Schürlialp (Latschüelfurgga) for average SWE of 180 mm (254 mm), and average spatial linear correlation with the reference SWE of 0.51 (0.48). However, the assimilated SWE observation must not be too early in the accumulation season or too late in the melt season when the snowpack is starting or ending to accumulate or melt, respectively. Smaller distances between ground tracks (1500 m and 500 m) show improved performance of the IES-FNN approach in space, with no significant improvement in terms of temporal reconstruction.</div><div>Applying the IES-FNN approach to e.g., real ICESat-2 data, remains challenging due to the higher uncertainties associated with these data. However, the approach we propose remains potentially very helpful in addressing the problem of scarcity of ground observations of precipitation and snow in high-mountain regions.</div></div>","PeriodicalId":36948,"journal":{"name":"Journal of Hydrology X","volume":"25 ","pages":"Article 100190"},"PeriodicalIF":3.1000,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Hydrology X","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2589915524000208","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Our understanding of the impact of climate change on water availability and natural hazards in high-mountain regions is limited due to the spatial and temporal scarcity of ground observations of precipitation and snow. Freely available, satellite-based information about the snowpack is currently mainly limited to indirect measurements of snow-covered area or very coarse-scale snow water equivalent (SWE), but only for flat areas in lowlands without vegetation cover. Novel space-based laser altimeters, such as ICESat-2, have the potential to provide high-resolution snow depth data in worldwide mountain regions where no ground observations exist. However, these space-based laser altimeters come with spatial gaps between ground tracks, obtained without repetition at a give location. To overcome these drawbacks, here, we present a combined probabilistic data assimilation and deep learning approach to reconstruct spatio-temporal SWE from observations of snow depth along ground tracks, imitating ICESat-2 tracks in view of a potential future global application.
Our approach is based on assimilating SWE and snow cover information in a degree-day model with an iterative ensemble smoother (IES) which allows temporally reconstructing SWE along hypothetical ground tracks separated by 3 km. As input, the degree-day model uses daily precipitation and downscaled air temperature from the ERA5 reanalysis. A feedforward neural network (FNN) is then used for spatial propagation of the daily mean and standard deviation of the updated SWE ensemble members obtained from the IES. The combined IES-FNN approach provides uncertainty-aware spatio-temporally continuous estimates of SWE.
We tested our approach in the alpine Dischma valley (Switzerland) using high-resolution snow depth maps obtained from photogrammetric techniques mounted on airplanes and unmanned aerial system observations. Our results show that the IES-FNN model provides reliable estimates at a resolution of approximately 100 m. Even assimilating only one SWE observation during the year (combined with satellite-based melt-out date estimates) produces satisfying results when evaluating the IES-FNN SWE reconstructions on independent dates and smaller (<4 km2) areas: mean absolute error of 86 mm (78 mm) at Schürlialp (Latschüelfurgga) for average SWE of 180 mm (254 mm), and average spatial linear correlation with the reference SWE of 0.51 (0.48). However, the assimilated SWE observation must not be too early in the accumulation season or too late in the melt season when the snowpack is starting or ending to accumulate or melt, respectively. Smaller distances between ground tracks (1500 m and 500 m) show improved performance of the IES-FNN approach in space, with no significant improvement in terms of temporal reconstruction.
Applying the IES-FNN approach to e.g., real ICESat-2 data, remains challenging due to the higher uncertainties associated with these data. However, the approach we propose remains potentially very helpful in addressing the problem of scarcity of ground observations of precipitation and snow in high-mountain regions.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
从稀疏地面轨迹重建连续时空 SWE 的数据同化与深度学习相结合方法
由于缺乏对降水和积雪的时空地面观测,我们对气候变化对高山地区水资源供应和自然灾害的影响的了解十分有限。目前,基于卫星的免费积雪信息主要限于间接测量积雪覆盖面积或非常粗略的雪水当量(SWE),但仅限于低地无植被覆盖的平坦区域。新型天基激光测高仪(如 ICESat-2)有可能在没有地面观测数据的全球山区提供高分辨率雪深数据。然而,这些天基激光测高仪的地面轨迹之间存在空间差距,在特定地点获得的数据不重复。为了克服这些缺点,我们在此提出了一种概率数据同化和深度学习相结合的方法,以模仿 ICESat-2 的轨迹,根据沿地面轨迹的雪深观测数据重建时空 SWE,从而在未来实现潜在的全球应用。我们的方法基于将 SWE 和雪盖信息同化到一个度日模型中,并使用迭代集合平滑器(IES),从而可以沿相距 3 公里的假定地面轨迹重建 SWE。度日模型使用ERA5再分析的日降水量和降尺度气温作为输入。然后使用前馈神经网络(FNN)对从 IES 中获得的 SWE 更新集合成员的日平均值和标准偏差进行空间传播。我们使用安装在飞机上的摄影测量技术和无人机系统观测所获得的高分辨率雪深图,在瑞士高山迪施玛山谷测试了我们的方法。结果表明,IES-FNN 模型可在约 100 米的分辨率范围内提供可靠的估计值。在评估独立日期和较小(4 平方公里)区域的 IES-FNN SWE 重建时,即使只同化一年中的一次 SWE 观测(结合基于卫星的融化日期估计),也能得出令人满意的结果:平均 SWE 为 180 毫米(254 毫米)时,Schürlialp(Latschüelfurgga)的平均绝对误差为 86 毫米(78 毫米),与参考 SWE 的平均空间线性相关为 0.51(0.48)。不过,同化的 SWE 观测值不能在积雪开始或结束积雪或融化的季节过早或过晚进行。地面轨迹之间的距离越小(1500 米和 500 米),IES-FNN 方法的空间性能就越好,但在时间重建方面没有明显改善。将 IES-FNN 方法应用于 ICESat-2 等真实数据仍具有挑战性,因为这些数据的不确定性更高。不过,我们提出的方法仍有可能非常有助于解决高山地区降水和降雪地面观测资料匮乏的问题。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Journal of Hydrology X
Journal of Hydrology X Environmental Science-Water Science and Technology
CiteScore
7.00
自引率
2.50%
发文量
20
审稿时长
25 weeks
期刊最新文献
Revisiting model complexity: Space-time correction of high dimensional variable sets in climate model simulations Quantifying the economic value of a national hydrometric network for households Private sensors and crowdsourced rainfall data: Accuracy and potential for modelling pluvial flooding in urban areas of Oslo, Norway A combined data assimilation and deep learning approach for continuous spatio-temporal SWE reconstruction from sparse ground tracks A nonstationary stochastic simulator for clustered regional hydroclimatic extremes to Characterize compound flood risk
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1