Regional Frequency Analysis at Ungauged Sites with Multivariate Adaptive Regression Splines

Msilini, Amina; Masselot, Pierre; Ouarda, Taha B. M. J.

doi:10.1175/jhm-d-19-0213.1

Cited by 13 publications

(3 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MARS model can be seen as a flexible extension of GAM, expressed as a linear combination of BFs and their interactions (Msilini et al, 2020). It is a flexible nonparametric regression approach that can deal with high‐dimensional data.…”

Section: Methodsmentioning

confidence: 99%

“…Ouarda et al (2022) used GAM also and temperature‐duration‐curves to develop a method for estimating river water temperature at ungauged sites, which they tested on 126 river thermal stations in eastern Canada and found to be robust and more accurate than a simple averaging method. The MARS approach has also been applied in various contexts, including the modeling of daily lake surface water temperature (Heddam et al, 2020), regional flood frequency analysis (Msilini et al, 2020) and streamflow simulation (Al‐Sudani et al, 2019). Saadi et al (2022) also used MARS, in addition to GAM, to simulate daily mean PTRA.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine‐learning modeling of hourly potential thermal refuge area: A case study from the Sainte‐Marguerite River (Quebec, Canada)

Hani

St‐Hilaire

Ouarda

2023

River Research & Apps

View full text Add to dashboard Cite

To understand the temporal and spatial variability of thermal refuges, this study focused on modeling potential thermal refuge area (PTRA) at a sub‐daily time‐step in two tributary confluences of the Sainte‐Marguerite River (Canada) during the summers of 2020 and 2021. Aquatic ectotherm species, such as Atlantic salmon (Salmo salar), seek these refuges to avoid heat stress during high summer river temperatures. To investigate the temporal variability of these PTRA, we employed inverse weighted distance interpolation to delineate the hourly area available at both confluences. We then analyzed the impact of the atypical low flow conditions of summer 2021 on the diel cycle of PTRA extremes using the coefficient of variation and the generalized additive model (GAM). Finally, we used four supervised machine‐learning regression models and three to five hydrometeorological predictors to estimate hourly PTRA availability: multivariate adaptive splines regression (MARS), GAM, support vector machine regression (SVM), and random forest regression (RF). The results showed that tree‐based and kernel‐based regression models, RF and SVM, outperformed GAM and MARS. RF had the highest accuracy at both sites, with a relative root mean square error and Nash–Sutcliffe efficiency coefficient (Nash) of 13% and 93%, respectively. Our study discovered that under warm conditions in August 2021, small perennial tributary inflows in combination with low mainstem discharge could create high and constant PTRA at confluences, potentially providing vital thermal refuges for cold‐water taxa. These refuges may be especially important at the local level, within a specific stretch or section of the river. Given the decreasing availability of thermal refuges for salmonids, it is crucial to monitor stream temperatures at small spatial and temporal scales using data‐driven techniques in order to understand stream temperature heterogeneity at tributary confluences.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Machine‐learning modeling of hourly potential thermal refuge area: A case study from the Sainte‐Marguerite River (Quebec, Canada)

Hani

St‐Hilaire

Ouarda

2023

River Research & Apps

View full text Add to dashboard Cite

show abstract

“…On the other hand, the execution speed of the MARS model is very high and there is no need to spend so much time. The MARS model has high accuracy in simulating PCP (Abraham et al 2001), flow discharge (Kisi et al 2022), suspended sediment load (Esmaeili-Gisavandani et al 2022) and flood (Msilini et al 2020).…”

Section: Multivariate Adaptive Regression Splines (Mars)mentioning

confidence: 99%

Improving the accuracy of satellite and reanalysis precipitation data by their ensemble usage

2022

View full text Add to dashboard Cite

This study evaluated the outputs of five precipitation (PCP) datasets. These models are ECMWF reanalysis 5th generation (ERA5), precipitation estimation from remotely sensed information using artificial neural networks-climate data record (PERSIANN-CDR), Asian precipitation-highly resolved observational data integration toward evaluation (APHRODITE), The national centers for environmental prediction climate forecast system reanalysis (NCEP CFSR) and climatic research unit (CRU). The PCP outputs of these models were compared with data of nine synoptic stations in the Khuzestan province. The results indicated a better match between the APHRODITE outputs and the PCP data at most stations (R2 > 0.85, root-mean-square error (RMSE) < 17.049 mm and − 4.25 < Bias < 2.633 mm). However, CRU model has the highest critical success index (more than 0.711) and the lowest false alarm ratio (less than 0.2) and ERA5 has the highest probability of detection (more than 0.967) at most stations. Then, PCP outputs of five reanalysis (ERA5), interpolated (APHRODITE, NCEP CFSR and CRU) and satellite (PERSIANN-CDR) PCP datasets were combined to reduce the PCP estimation error. The multivariate adaptive regression splines models were employed for this purpose. The results show that the RMSE of all the stations, except Ahvaz station, decreased and the BIAS decreased too. Given the results, using ensemble data methods is a suitable way for reducing the error and increasing the accuracy of these models.

show abstract

Evaluation of additional physiographical variables characterising drainage network systems in regional frequency analysis, a Quebec watersheds case-study

Msilini¹,

Ouarda²,

Masselot

2021

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Regional Frequency Analysis (RFA) relies on a wide range of physiographical and meteorological variables to estimate hydrological quantiles at ungauged sites. However, additional catchment characteristics related to its drainage network are not yet fully understood and integrated in RFA procedures. The aim of the present paper is to propose the integration of several physiographical variables characterizing the drainage network systems in RFA, and to evaluate their added value in predicting quantiles at ungauged sites. The proposed extended dataset (EXTD) includes several variables characterising drainage network characteristics. To evaluate the new variables, a number of commonly used RFA approaches are applied to the extended data representing 151 stations in Quebec (Canada) and compared to a standard dataset (STA) that excludes the new variables. The considered RFA approaches include the combination of two neighborhood methods namely the canonical correlation analysis (CCA) and the region of influence (ROI) with two regional estimation (RE) models which are the log-linear regression model (LLRM) and the generalized additive model (GAM). The RE models are also applied without the hydrological neighborhood. Results show that regional models using the extended dataset lead to significantly better flood quantile predictions, especially for large basins. Indeed, the variable selection performed with EXTD consistently includes some of the new variables, in particular the drainage density, the stream length ratio, and the ruggedness number. Two other new variables are also identified and included in the DHR step: the circularity ratio and the texture ratio. This leads to better predictions with relative errors about 29% for EXTD, versus around 42% for STA in the case of the best

show abstract

Regional Frequency Analysis at Ungauged Sites with Multivariate Adaptive Regression Splines

Cited by 13 publications

References 69 publications

Machine‐learning modeling of hourly potential thermal refuge area: A case study from the Sainte‐Marguerite River (Quebec, Canada)

Machine‐learning modeling of hourly potential thermal refuge area: A case study from the Sainte‐Marguerite River (Quebec, Canada)

Improving the accuracy of satellite and reanalysis precipitation data by their ensemble usage

Evaluation of additional physiographical variables characterising drainage network systems in regional frequency analysis, a Quebec watersheds case-study

Contact Info

Product

Resources

About