DOI: 10.5194/hess-24-4887-2020
论文题名: Application of machine learning techniques for regional bias correction of snow water equivalent estimates in Ontario; Canada
作者: King F. ; Erler A.R. ; Frey S.K. ; Fletcher C.G.
刊名: Hydrology and Earth System Sciences
ISSN: 1027-5606
出版年: 2020
卷: 24, 期: 10 起始页码: 4887
结束页码: 4902
语种: 英语
Scopus关键词: Budget control
; Decision trees
; Flood control
; Floods
; Machine learning
; Snow melting systems
; Surveys
; Water management
; Data assimilation systems
; Interannual variability
; Machine learning methods
; Machine learning techniques
; Snow water equivalent
; Spatio-temporal domains
; Statistical techniques
; Waterresource management
; Snow
; correction
; discharge
; flood forecasting
; hydrograph
; machine learning
; nonlinearity
; resource management
; sampling bias
; snow water equivalent
; snowmelt
; snowpack
; Great Lakes [North America]
英文摘要: Snow is a critical contributor to Ontario's waterenergy budget, with impacts on water resource management and flood forecasting. Snow water equivalent (SWE) describes the amount of water stored in a snowpack and is important in deriving estimates of snowmelt. However, only a limited number of sparsely distributed snow survey sites (n = 383) exist throughout Ontario. The SNOw Data Assimilation System (SNODAS) is a daily, 1 km gridded SWE product that provides uniform spatial coverage across this region; however, we show here that SWE estimates from SNODAS display a strong positive mean bias of 50% (16mm SWE) when compared to in situ observations from 2011 to 2018. This study evaluates multiple statistical techniques of varying complexity, including simple subtraction, linear regression and machine learning methods to bias-correct SNODAS SWE estimates using absolute mean bias and RMSE as evaluation criteria. Results show that the random forest (RF) algorithm is most effective at reducing bias in SNODAS SWE, with an absolute mean bias of 0.2 mm and RMSE of 3.64 mm when compared with in situ observations. Other methods, such as mean bias subtraction and linear regression, are somewhat effective at bias reduction; however, only the RF method captures the nonlinearity in the bias and its interannual variability. Applying the RF model to the full spatio-temporal domain shows that the SWE bias is largest before 2015, during the spring melt period, north of 44.5° N and east (downwind) of the Great Lakes. As an independent validation, we also compare estimated snowmelt volumes with observed hydrographs and demonstrate that uncorrected SNODAS SWE is associated with unrealistically large volumes at the time of the spring freshet, while bias-corrected SWE values are highly consistent with observed discharge volumes. © 2020 Now Publishers Inc. All rights reserved. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/162574
Appears in Collections: 气候变化与战略
There are no files associated with this item.
作者单位: King, F., Deptartment of Geography & Environmental Management, University of WaterlooON, Canada; Erler, A.R., Aquanty, Waterloo, ON, Canada; Frey, S.K., Aquanty, Waterloo, ON, Canada, Deptartment of Earth & Environmental Sciences, University of WaterlooON, Canada; Fletcher, C.G., Deptartment of Geography & Environmental Management, University of WaterlooON, Canada
Recommended Citation:
King F.,Erler A.R.,Frey S.K.,et al. Application of machine learning techniques for regional bias correction of snow water equivalent estimates in Ontario; Canada[J]. Hydrology and Earth System Sciences,2020-01-01,24(10)