scholarly journals Enhanced streamflow prediction with SWAT using support vector regression for spatial calibration: A case study in the Illinois River watershed, U.S.

PLoS ONE ◽  
2021 ◽  
Vol 16 (4) ◽  
pp. e0248489
Author(s):  
Lifeng Yuan ◽  
Kenneth J. Forshay
Keyword(s):  
Support Vector Regression ◽  
Assessment Tool ◽  
Swat Model ◽  
Illinois River ◽  
Support Vector ◽  
Streamflow Prediction ◽  
River Watershed ◽  
Monthly Streamflow ◽  
Non Linear ◽  
Spatial Calibration

Accurate streamflow prediction plays a pivotal role in hydraulic project design, nonpoint source pollution estimation, and water resources planning and management. However, the highly non-linear relationship between rainfall and runoff makes prediction difficult with desirable accuracy. To improve the accuracy of monthly streamflow prediction, a seasonal Support Vector Regression (SVR) model coupled to the Soil and Water Assessment Tool (SWAT) model was developed for 13 subwatersheds in the Illinois River watershed (IRW), U.S. Terrain, precipitation, soil, land use and land cover, and monthly streamflow data were used to build the SWAT model. SWAT Streamflow output and the upstream drainage area were used as two input variables into SVR to build the hybrid SWAT-SVR model. The Calibration Uncertainty Procedure (SWAT-CUP) and Sequential Uncertainty Fitting-2 (SUFI-2) algorithms were applied to compare the model performance against SWAT-SVR. The spatial calibration and leave-one-out sampling methods were used to calibrate and validate the hybrid SWAT-SVR model. The results showed that the SWAT-SVR model had less deviation and better performance than SWAT-CUP simulations. SWAT-SVR predicted streamflow more accurately during the wet season than the dry season. The model worked well when it was applied to simulate medium flows with discharge between 5 m3 s-1 and 30 m3 s-1, and its applicable spatial scale fell between 500 to 3000 km2. The overall performance of the model on yearly time series is “Satisfactory”. This new SWAT-SVR model has not only the ability to capture intrinsic non-linear behaviors between rainfall and runoff while considering the mechanism of runoff generation but also can serve as a reliable regional tool for an ungauged or limited data watershed that has similar hydrologic characteristics with the IRW.

scholarly journals Hydrological simulation in a tropical humid basin in the Cerrado biome using the SWAT model

2018 ◽  
Vol 49 (3) ◽  
pp. 908-923 ◽  
Author(s):  
Richarde Marques da Silva ◽  
José Carlos Dantas ◽  
Joyce de Araújo Beltrão ◽  
Celso A. G. Santos
Keyword(s):  
Assessment Tool ◽  
Swat Model ◽  
Coefficient Of Determination ◽  
Southeastern Brazil ◽  
Measured Variable ◽  
Hydrological Simulation ◽  
Cerrado Biome ◽  
Monthly Streamflow ◽  
P Factor ◽  
Water Assessment

Abstract A Soil and Water Assessment Tool (SWAT) model was used to model streamflow in a tropical humid basin in the Cerrado biome, southeastern Brazil. This study was undertaken in the Upper São Francisco River basin, because this basin requires effective management of water resources in drought and high-flow periods. The SWAT model was calibrated for the period of 1978–1998 and validated for 1999–2007. To assess the model calibration and uncertainty, four indices were used: (a) coefficient of determination (R2); (b) Nash–Sutcliffe efficiency (NS); (c) p-factor, the percentage of data bracketed by the 95% prediction uncertainty (95PPU); and (d) r-factor, the ratio of average thickness of the 95PPU band to the standard deviation of the corresponding measured variable. In this paper, average monthly streamflow from three gauges (Porto das Andorinhas, Pari and Ponte da Taquara) were used. The results indicated that the R2 values were 0.73, 0.80 and 0.76 and that the NS values were 0.68, 0.79 and 0.73, respectively, during the calibration. The validation also indicated an acceptable performance with R2 = 0.80, 0.76, 0.60 and NS = 0.61, 0.64 and 0.58, respectively. This study demonstrates that the SWAT model provides a satisfactory tool to assess basin streamflow and management in Brazil.

scholarly journals Performance of NO, NO<sub>2</sub> low cost sensors and three calibration approaches within a real world application

2018 ◽  
Vol 11 (6) ◽  
pp. 3717-3735 ◽  
Author(s):  
Alessandro Bigi ◽  
Michael Mueller ◽  
Stuart K. Grange ◽  
Grazia Ghermandi ◽  
Christoph Hueglin
Keyword(s):  
Random Forest ◽  
Linear Regression ◽  
Support Vector Regression ◽  
Low Cost ◽  
Multivariate Linear Regression ◽  
Pollution Level ◽  
Support Vector ◽  
Non Linear ◽  
Linear Algorithms ◽  
Urban Sites

Abstract. Low cost sensors for measuring atmospheric pollutants are experiencing an increase in popularity worldwide among practitioners, academia and environmental agencies, and a large amount of data by these devices are being delivered to the public. Notwithstanding their behaviour, performance and reliability are not yet fully investigated and understood. In the present study we investigate the medium term performance of a set of NO and NO2 electrochemical sensors in Switzerland using three different regression algorithms within a field calibration approach. In order to mimic a realistic application of these devices, the sensors were initially co-located at a rural regulatory monitoring site for a 4-month calibration period, and subsequently deployed for 4 months at two distant regulatory urban sites in traffic and urban background conditions, where the performance of the calibration algorithms was explored. The applied algorithms were Multivariate Linear Regression, Support Vector Regression and Random Forest; these were tested, along with the sensors, in terms of generalisability, selectivity, drift, uncertainty, bias, noise and suitability for spatial mapping intra-urban pollution gradients with hourly resolution. Results from the deployment at the urban sites show a better performance of the non-linear algorithms (Support Vector Regression and Random Forest) achieving RMSE  <  5 ppb, R2 between 0.74 and 0.95 and MAE between 2 and 4 ppb. The combined use of both NO and NO2 sensor output in the estimate of each pollutant showed some contribution by NO sensor to NO2 estimate and vice-versa. All algorithms exhibited a drift ranging between 5 and 10 ppb for Random Forest and 15 ppb for Multivariate Linear Regression at the end of the deployment. The lowest concentration correctly estimated, with a 25 % relative expanded uncertainty, resulted in ca. 15–20 ppb and was provided by the non-linear algorithms. As an assessment for the suitability of the tested sensors for a targeted application, the probability of resolving hourly concentration difference in cities was investigated. It was found that NO concentration differences of 5–10 ppb (8–10 for NO2) can reliably be detected (90 % confidence), depending on the air pollution level. The findings of this study, although derived from a specific sensor type and sensor model, are based on a flexible methodology and have extensive potential for exploring the performance of other low cost sensors, that are different in their target pollutant and sensing technology.

scholarly journals Predicting Rainfall and Runoff Through Satellite Soil Moisture Data and SWAT Modelling for a Poorly Gauged Basin in Iran

Water ◽  
2019 ◽  
Vol 11 (3) ◽  
pp. 594 ◽  
Author(s):  
Majid Fereidoon ◽  
Manfred Koch ◽  
Luca Brocca
Keyword(s):  
Soil Moisture ◽  
Hydrological Model ◽  
Assessment Tool ◽  
Swat Model ◽  
Rainfall Data ◽  
Coefficient Of Determination ◽  
Hydrological Models ◽  
Streamflow Prediction ◽  
Soil Moisture Data ◽  
Monthly Runoff

Hydrological models are widely used for many purposes in water sector projects, including streamflow prediction and flood risk assessment. Among the input data used in such hydrological models, the spatial-temporal variability of rainfall datasets has a significant role on the final discharge estimation. Therefore, accurate measurements of rainfall are vital. On the other hand, ground-based measurement networks, mainly in developing countries, are either nonexistent or too sparse to capture rainfall accurately. In addition to in-situ rainfall datasets, satellite-derived rainfall products are currently available globally with high spatial and temporal resolution. An innovative approach called SM2RAIN that estimates rainfall from soil moisture data has been applied successfully to various regions. In this study, first, soil moisture content derived from the Advanced Microwave Scanning Radiometer for the Earth observing system (AMSR-E) is used as input into the SM2RAIN algorithm to estimate daily rainfall (SM2R-AMSRE) at different sites in the Karkheh river basin (KRB), southwest Iran. Second, the SWAT (Soil and Water Assessment Tool) hydrological model was applied to simulate runoff using both ground-based observed rainfall and SM2R-AMSRE rainfall as input. The results reveal that the SM2R-AMSRE rainfall data are, in most cases, in good agreement with ground-based rainfall, with correlations R ranging between 0.58 and 0.88, though there is some underestimation of the observed rainfall due to soil moisture saturation not accounted for in the SM2RAIN equation. The subsequent SWAT-simulated monthly runoff from SM2R-AMSRE rainfall data (SWAT-SM2R-AMSRE) reproduces the observations at the six gauging stations (with coefficient of determination, R² > 0.71 and NSE > 0.56), though with slightly worse performances in terms of bias (Bias) and root-mean-square error (RMSE) and, again, some systematic flow underestimation compared to the SWAT model with ground-based rainfall input. Additionally, rainfall estimates of two satellite products of the Tropical Rainfall Measuring Mission (TRMM), 3B42 and 3B42RT, are used in the calibrated SWAT- model after bias correction. The monthly runoff predictions obtained with 3B42- rainfall have 0.42 < R2 < 0.72 and−0.06 < NSE < 0.74 which are slightly better than those obtained with 3B42RT- rainfall, but not as good as the SWAT-SM2R-AMSRE. Therefore, despite the aforementioned limitations, using SM2R-AMSRE rainfall data in a hydrological model like SWAT appears to be a viable approach in basins with limited ground-based rainfall data.

Sign in / Sign up

Export Citation Format

Share Document