scholarly journals Alternative configurations of quantile regression for estimating predictive uncertainty in water level forecasts for the upper Severn River: a comparison

2014 ◽  
Vol 18 (9) ◽  
pp. 3411-3428 ◽  
Author(s):  
P. López López ◽  
J. S. Verkade ◽  
A. H. Weerts ◽  
D. P. Solomatine
Keyword(s):  
Quantile Regression ◽  
Water Level ◽  
Skill Score ◽  
Water Levels ◽  
Normal Space ◽  
Numerical Verification ◽  
Relative Operating Characteristic ◽  
Hydrological Uncertainty ◽  
Quality Aspects ◽  
Normal Quantile Transformation

Abstract. The present study comprises an intercomparison of different configurations of a statistical post-processor that is used to estimate predictive hydrological uncertainty. It builds on earlier work by Weerts, Winsemius and Verkade (2011; hereafter referred to as WWV2011), who used the quantile regression technique to estimate predictive hydrological uncertainty using a deterministic water level forecast as a predictor. The various configurations are designed to address two issues with the WWV2011 implementation: (i) quantile crossing, which causes non-strictly rising cumulative predictive distributions, and (ii) the use of linear quantile models to describe joint distributions that may not be strictly linear. Thus, four configurations were built: (i) a ''classical" quantile regression, (ii) a configuration that implements a non-crossing quantile technique, (iii) a configuration where quantile models are built in normal space after application of the normal quantile transformation (NQT) (similar to the implementation used by WWV2011), and (iv) a configuration that builds quantile model separately on separate domains of the predictor. Using each configuration, four reforecasting series of water levels at 14 stations in the upper Severn River were established. The quality of these four series was intercompared using a set of graphical and numerical verification metrics. Intercomparison showed that reliability and sharpness vary across configurations, but in none of the configurations do these two forecast quality aspects improve simultaneously. Further analysis shows that skills in terms of the Brier skill score, mean continuous ranked probability skill score and relative operating characteristic score is very similar across the four configurations.

scholarly journals Alternative configurations of Quantile Regression for estimating predictive uncertainty in water level forecasts for the Upper Severn River: a comparison

2014 ◽  
Vol 11 (4) ◽  
pp. 3811-3855 ◽  
Author(s):  
P. López López ◽  
J. S. Verkade ◽  
A. H. Weerts ◽  
D. P. Solomatine
Keyword(s):  
Quantile Regression ◽  
Water Level ◽  
Skill Score ◽  
Water Levels ◽  
Normal Space ◽  
Numerical Verification ◽  
Relative Operating Characteristic ◽  
Hydrological Uncertainty ◽  
Is Implementation ◽  
Quality Aspects

Abstract. The present study comprises an inter-comparison of different configurations of a statistical post-processor that is used to estimate predictive hydrological uncertainty. It builds on earlier work by Weerts et al. (2011, herinafter referred to as wwv2011), who used the Quantile Regression technique to estimate predictive hydrological uncertainty using a deterministic water level forecast as a predictor. The various configurations are designed to address two issues with the wwv2011 implementation: (i) quantile crossing, which causes non-strictly rising cumulative predictive distributions, and (ii) the use of linear quantile models to describe joint distributions that may not be strictly linear. Thus, four configurations were built: (i) the "as is" implementation used by wwv2011, (ii) a configuration that implements a non-crossing quantile technique, (iii) a configuration where quantile models are built in Normal space after application of the Normal Quantile Transform, and (iv) a configuration that builds quantile model separately on separate domains of the predictor. Using each, four re-forecasting series of water levels at fourteen stations in the Upper Severn River were established. The quality of these four series was inter-compared using a set of graphical and numerical verification metrics. Intercomparison showed that reliability and sharpness vary across configurations, but in none of the configurations do these two forecast quality aspects improve simultaneously. Further analysis shows that skills in terms of Brier Skill Score, mean Continuous Ranked Probability Skill Score and Relative Operating Characteristic Score is very similar across the four configurations.

scholarly journals Performance and robustness of probabilistic river forecasts computed with quantile regression based on multiple independent variables

2015 ◽  
Vol 19 (9) ◽  
pp. 3969-3990 ◽  
Author(s):  
F. Hoss ◽  
P. S. Fischbeck
Keyword(s):  
Quantile Regression ◽  
High Reliability ◽  
Forecast Error ◽  
Skill Score ◽  
Water Levels ◽  
Training Data ◽  
Forecast Errors ◽  
Data Set ◽  
Forecast Performance ◽  
Independent Variables

Abstract. This study applies quantile regression (QR) to predict exceedance probabilities of various water levels, including flood stages, with combinations of deterministic forecasts, past forecast errors and rates of water level rise as independent variables. A computationally cheap technique to estimate forecast uncertainty is valuable, because many national flood forecasting services, such as the National Weather Service (NWS), only publish deterministic single-valued forecasts. The study uses data from the 82 river gauges, for which the NWS' North Central River Forecast Center issues forecasts daily. Archived forecasts for lead times of up to 6 days from 2001 to 2013 were analyzed. Besides the forecast itself, this study uses the rate of rise of the river stage in the last 24 and 48 h and the forecast error 24 and 48 h ago as predictors in QR configurations. When compared to just using the forecast as an independent variable, adding the latter four predictors significantly improved the forecasts, as measured by the Brier skill score and the continuous ranked probability score. Mainly, the resolution increases, as the forecast-only QR configuration already delivered high reliability. Combining the forecast with the other four predictors results in a much less favorable performance. Lastly, the forecast performance does not strongly depend on the size of the training data set but on the year, the river gauge, lead time and event threshold that are being forecast. We find that each event threshold requires a separate configuration or at least calibration.

scholarly journals Estimation of predictive hydrological uncertainty using quantile regression: examples from the National Flood Forecasting System (England and Wales)

2011 ◽  
Vol 15 (1) ◽  
pp. 255-265 ◽  
Author(s):  
A. H. Weerts ◽  
H. C. Winsemius ◽  
J. S. Verkade
Keyword(s):  
Quantile Regression ◽  
Water Level ◽  
Flood Forecasting ◽  
Water Levels ◽  
Forecast Errors ◽  
Rainfall Runoff ◽  
England And Wales ◽  
Predictive Uncertainty ◽  
Hydrological Characteristics ◽  
Forecasting System

Abstract. In this paper, a technique is presented for assessing the predictive uncertainty of rainfall-runoff and hydraulic forecasts. The technique conditions forecast uncertainty on the forecasted value itself, based on retrospective Quantile Regression of hindcasted water level forecasts and forecast errors. To test the robustness of the method, a number of retrospective forecasts for different catchments across England and Wales having different size and hydrological characteristics have been used to derive in a probabilistic sense the relation between simulated values of water levels and matching errors. From this study, we can conclude that using Quantile Regression for estimating forecast errors conditional on the forecasted water levels provides a relatively simple, efficient and robust means for estimation of predictive uncertainty.

scholarly journals Estimation of predictive hydrological uncertainty using quantile regression: examples from the national flood forecasting system (England and Wales)

2010 ◽  
Vol 7 (4) ◽  
pp. 5547-5575 ◽  
Author(s):  
A. H. Weerts ◽  
H. C. Winsemius ◽  
J. S. Verkade
Keyword(s):  
Quantile Regression ◽  
Water Level ◽  
Uncertainty Estimation ◽  
Water Levels ◽  
Forecast Errors ◽  
Rainfall Runoff ◽  
England And Wales ◽  
Predictive Uncertainty ◽  
Hydrological Characteristics ◽  
Forecasting System

Abstract. In this paper, a is presented for assessing the predictive uncertainty of rainfall-runoff and hydraulic forecasts that conditions forecasts uncertainty on the forecasted value itself, based on retrospective quantile regression of hindcasted water level forecasts and forecast errors. To test the robustness of the method, a number of retrospective forecasts for different catchments across England and Wales having different size and hydrological characteristics have been used to derive in a probabilistic sense the relation between simulated values of discharges and water levels, and matching errors. From this study, we can conclude that using quantile regression for estimating forecast errors conditional on the forecasted water levels provides an extremely simple, efficient and robust means for uncertainty estimation of deterministic forecasts.

AUTOMATIC WATER LEVEL MONITORING WITH IOT AND LPWAN

2019 ◽  
pp. 95-103
Author(s):  
Krum Videnov ◽  
Vanya Stoykova
Keyword(s):  
Water Level ◽  
Real Time ◽  
Early Warning ◽  
Water Sources ◽  
Water Levels ◽  
Water Level Fluctuations ◽  
Level Monitoring ◽  
Water Level Monitoring

Monitoring water levels of lakes, streams, rivers and other water basins is of essential importance and is a popular measurement for a number of different industries and organisations. Remote water level monitoring helps to provide an early warning feature by sending advance alerts when the water level is increased (reaches a certain threshold). The purpose of this report is to present an affordable solution for measuring water levels in water sources using IoT and LPWAN. The assembled system enables recording of water level fluctuations in real time and storing the collected data on a remote database through LoRaWAN for further processing and analysis.

Interaction between water pressure in the basal drainage system and discharge from an Alpine glacier before and during a rainfall-induced subglacial hydrological event

1997 ◽  
Vol 24 ◽  
pp. 288-292 ◽  
Author(s):  
Andrew P. Barrett ◽  
David N. Collins
Keyword(s):  
Water Level ◽  
Water Pressure ◽  
Drainage System ◽  
Water Levels ◽  
Drainage Network ◽  
Alpine Glacier ◽  
Hydraulic Conditions ◽  
Borehole Water ◽  
Progressive Growth ◽  
Resultant Increase

Combined measurements of meltwater discharge from the portal and of water level in a borehole drilled to the bed of Findelengletscher, Switzerland, were obtained during the later part of the 1993 ablation season. A severe storm, lasting from 22 through 24 September, produced at least 130 mm of precipitation over the glacier, largely as rain. The combined hydrological records indicate periods during which the basal drainage system became constricted and water storage in the glacier increased, as well as phases of channel growth. During the storm, water pressure generally increased as water backed up in the drainage network. Abrupt, temporary falls in borehole water level were accompanied by pulses in portal discharge. On 24 September, whilst borehole water level continued to rise, water started to escape under pressure with a resultant increase in discharge. As the drainage network expanded, a large amount of debris was flushed from a wide area of the bed. Progressive growth in channel capacity as discharge increased enabled stored water to drain and borehole water level to fall rapidly. Possible relationships between observed borehole water levels and water pressures in subglacial channels are influenced by hydraulic conditions at the base of the hole, distance between the hole and a channel, and the nature of the substrate.

scholarly journals Spatial distribution of water level impact to back-barrier bays

2018 ◽  
Author(s):  
Alfredo L. Aretxabaleta ◽  
Neil K. Ganju ◽  
Zafer Defne ◽  
Richard P. Signell
Keyword(s):  
Water Level ◽  
Sea Level ◽  
Water Levels ◽  
Analytical Models ◽  
Barnegat Bay ◽  
Sea Level Fluctuations ◽  
Flooding Hazard ◽  
Short Period ◽  
Inlet Geometry ◽  
Spatial Estimates

Abstract. Water level in semi-enclosed bays, landward of barrier islands, is mainly driven by offshore sea level fluctuations that are modulated by bay geometry and bathymetry, causing spatial variability in the ensuing response (transfer). Local wind setup can have a secondary role that depends on wind speed, fetch, and relative orientation of the wind direction and the bay. Inlet geometry and bathymetry primarily regulate the magnitude of the transfer between open ocean and bay. Tides and short-period offshore oscillations are more damped in the bays than longer-lasting offshore fluctuations, such as storm surge and sea level rise. We compare observed and modeled water levels at stations in a mid-Atlantic bay (Barnegat Bay) with offshore water level proxies. Observed water levels in Barnegat Bay are compared and combined with model results from the Coupled Ocean–Atmosphere–Wave–Sediment Transport (COAWST) modeling system to evaluate the spatial structure of the water level transfer. Analytical models based on the dimensional characteristics of the bay are used to combine the observed data and the numerical model results in a physically consistent approach. Model water level transfers match observed values at locations inside the Bay in the storm frequency band (transfers ranging from 70–100 %) and tidal frequencies (10–55 %). The contribution of frequency-dependent local setup caused by wind acting along the bay is also considered. The approach provides transfer estimates for locations inside the Bay where observations were not available resulting in a complete spatial characterization. The approach allows for the study of the Bay response to alternative forcing scenarios (landscape changes, future storms, and rising sea level). Detailed spatial estimates of water level transfer can inform decisions on inlet management and contribute to the assessment of current and future flooding hazard in back-barrier bays and along mainland shorelines.

Sign in / Sign up

Export Citation Format

Share Document