Multi-site multi-variable hydrologic model development for spatially heterogeneous river basins to achieve realistic basin modelling

2020 ◽  
Author(s):  
Saumya Srivastava ◽  
Nagesh Kumar Dasika
Keyword(s):  
Soil Moisture ◽  
Spatial Variability ◽  
Uncertainty Analysis ◽  
River Basin ◽  
Performance Measures ◽  
Model Calibration ◽  
Sediment Load ◽  
Hydrologic Model ◽  
Performance Measure ◽  
Hydrologic Models

<p>Hydrologic modelling is an indispensable tool for simulation of river basin processes in water resources planning and management. Hydrologic models are used to understand dynamic interactions between climate and river basin hydrology. Model calibration, validation, parameter sensitivity and uncertainty analysis are essential prior to the application of hydrologic models. A large catchment with high spatial variability and heterogeneity can be modeled realistically when calibration is done at multiple locations, for multiple hydrologic variables like streamflow, soil moisture, sediment flow, evapotranspiration, etc. This ensures maximum utilization of field measurements of the hydrological variables, reduces the uncertainty in parameter identification and highlights the areas that need greater calibration effort. In the present study, hydrologic model simulations are run for the Mahanadi river basin in India using SWAT (Soil and Water Assessment Tool) and model calibration, uncertainty analysis, sensitivity analysis and validation are performed using SUFI-2 optimization algorithm in SWAT-CUP (SWAT Calibration and Uncertainty Programs). Entire Mahanadi basin is calibrated for several variables like streamflow, soil moisture, sediment load and evapotranspiration at various locations. The spatial heterogeneity of the catchment is taken into account in model calibration by choosing appropriate ranges of different parameters for each sub basin based on the soil types, slope classes and land use land cover present in the sub basins. When multi-site multi-variable calibration is carried out, serial calibration for individual variables and locations gives different result when compared with the simultaneous calibration for all variables and locations. In this study, a comparison of serial calibration for individual hydrologic variables and calibration sites versus simultaneous calibration for all hydrologic variables and calibration sites is made. Various performance measures like Nash-Sutcliffe efficiency (NSE), percent bias, coefficient of determination, modified NSE, etc. are used to quantify the model fit between the observed and the simulated values of various variables. The choice of performance measure affects the calibration solution, and depends on the calibration variables for which observed data is available. The performances of the fitted parameters are conditional with respect to the calibration variables and the choice of the performance measure. The present study talks about the suitability of the performance measure to different hydrologic variables like streamflow, sediment load, soil moisture, etc. The model simulation results for the Mahanadi river basin are compared with the observed values of hydrologic variables using different performance measures for calibration and validation of the model. The results show that model performance is enhanced when it is calibrated at multiple locations, for multiple variables, by taking the spatial variability of parameters across various sub-basins into account. This study explores the suitability of different performance measures for different hydrologic variables and compares the serial and simultaneous calibration for multiple hydrologic variables at multiple locations.</p>

scholarly journals Spatial Pattern Oriented Multi-Criteria Sensitivity Analysis of a Distributed Hydrologic Model

2018 ◽  
Author(s):  
Mehmet Cüneyd Demirel ◽  
Julian Koch ◽  
Gorka Mendiguren ◽  
Simon Stisen
Keyword(s):  
Sensitivity Analysis ◽  
Spatial Variability ◽  
Spatial Pattern ◽  
Spatial Patterns ◽  
Performance Measures ◽  
Sampling Strategy ◽  
Behavioral Model ◽  
Hydrologic Model ◽  
Actual Evapotranspiration ◽  
Model Parameters

Hydrologic models are conventionally constrained and evaluated using point measurements of streamflow, which represents an aggregated catchment measure. As a consequence of this single objective focus, model parametrization and model parameter sensitivity are typically not reflecting other aspects of catchment behavior. Specifically for distributed models, the spatial pattern aspect is often overlooked. Our paper examines the utility of multiple performance measures in a spatial sensitivity analysis framework to determine the key parameters governing the spatial variability of predicted actual evapotranspiration (AET). Latin hypercube one-at-a-time (LHS-OAT) sampling strategy with multiple initial parameter sets was applied using the mesoscale hydrologic model (mHM) and a total of 17 model parameters were identified as sensitive. The results indicate different parameter sensitivities for different performance measures focusing on temporal hydrograph dynamics and spatial variability of actual evapotranspiration. While spatial patterns were found to be sensitive to vegetation parameters, streamflow dynamics were sensitive to pedo-transfer function (PTF) parameters. Above all, our results show that behavioral model definition based only on streamflow metrics in the generalized likelihood uncertainty estimation (GLUE) type methods require reformulation by incorporating spatial patterns into the definition of threshold values to reveal robust hydrologic behavior in the analysis.

scholarly journals On the scaling characteristics of observed and simulated spatial soil moisture fields

2009 ◽  
Vol 16 (1) ◽  
pp. 141-150 ◽  
Author(s):  
M. Gebremichael ◽  
R. Rigon ◽  
G. Bertoldi ◽  
T. M. Over
Keyword(s):  
Soil Moisture ◽  
Water Content ◽  
Soil Water ◽  
Soil Water Content ◽  
Spatial Correlation ◽  
Hydrologic Model ◽  
Invariance Property ◽  
Spatial Average ◽  
Hydrologic Models ◽  
Model Simulations

Abstract. By providing continuous high-resolution simulations of soil moisture fields, distributed hydrologic models could be powerful tools to advance the scientific community's understanding of the space-time variability and scaling characteristics of soil moisture fields. However, in order to use the soil moisture simulations from hydrologic models with confidence, it is important to understand whether the models are able to represent in a reliable way the processes regulating soil moisture variability. In this study, a comparison of the scaling characteristics of spatial soil moisture fields derived from a set of microwave radiometer observations from the Southern Great Plains 1997 experiment and corresponding simulations using the distributed hydrologic model GEOtop is performed through the use of generalized variograms. Microwave observations and model simulations are in agreement with respect to suggesting the existence of a scale-invariance property in the variograms of spatial soil moisture fields, and indicating that the scaling characteristics vary with changes in the spatial average soil water content. However, observations and simulations give contradictory results regarding the relationship between the scaling parameters (i.e. spatial organization) and average soil water content. The drying process increased the spatial correlation of the microwave observations at both short and long separation distances while increasing the rate of decay of correlation with distance. The effect of drying on the spatial correlation of the model simulations was more complex, depending on the storm and the simulation examined, but for the largest storm in the simulation most similar to the observations, drying increased the long-range correlation but decreased the short-range. This is an indication that model simulations, while reproducing correctly the total streamflow at the outlet of the watershed, may not accurately reproduce the runoff production mechanisms. Consideration of the scaling characteristics of spatial soil moisture fields can therefore serve as a more intensive means for validating distributed hydrologic models, compared to the traditional approach of only comparing the streamflow hydrographs.

scholarly journals Multi-site calibration and validation of SWAT with satellite-based evapotranspiration in a data-sparse catchment in southwestern Nigeria

2019 ◽  
Vol 23 (2) ◽  
pp. 1113-1144 ◽  
Author(s):  
Abolanle E. Odusanya ◽  
Bano Mehdi ◽  
Christoph Schürz ◽  
Adebayo O. Oke ◽  
Olufiropo S. Awokola ◽  
...  
Keyword(s):  
Soil Moisture ◽  
River Basin ◽  
Model Calibration ◽  
Hydrological Model ◽  
Assessment Tool ◽  
Potential Evapotranspiration ◽  
Swat Model ◽  
Water Balance Components ◽  
Southwestern Nigeria ◽  
Calibration And Validation

Abstract. The main objective of this study was to calibrate and validate the eco-hydrological model Soil and Water Assessment Tool (SWAT) with satellite-based actual evapotranspiration (AET) data from the Global Land Evaporation Amsterdam Model (GLEAM_v3.0a) and from the Moderate Resolution Imaging Spectroradiometer Global Evaporation (MOD16) for the Ogun River Basin (20 292 km2) located in southwestern Nigeria. Three potential evapotranspiration (PET) equations (Hargreaves, Priestley–Taylor and Penman–Monteith) were used for the SWAT simulation of AET. The reference simulations were the three AET variables simulated with SWAT before model calibration took place. The sequential uncertainty fitting technique (SUFI-2) was used for the SWAT model sensitivity analysis, calibration, validation and uncertainty analysis. The GLEAM_v3.0a and MOD16 products were subsequently used to calibrate the three SWAT-simulated AET variables, thereby obtaining six calibrations–validations at a monthly timescale. The model performance for the three SWAT model runs was evaluated for each of the 53 subbasins against the GLEAM_v3.0a and MOD16 products, which enabled the best model run with the highest-performing satellite-based AET product to be chosen. A verification of the simulated AET variable was carried out by (i) comparing the simulated AET of the calibrated model to GLEAM_v3.0b AET, which is a product that has different forcing data than the version of GLEAM used for the calibration, and (ii) assessing the long-term average annual and average monthly water balances at the outlet of the watershed. Overall, the SWAT model, composed of the Hargreaves PET equation and calibrated using the GLEAM_v3.0a data (GS1), performed well for the simulation of AET and provided a good level of confidence for using the SWAT model as a decision support tool. The 95 % uncertainty of the SWAT-simulated variable bracketed most of the satellite-based AET data in each subbasin. A validation of the simulated soil moisture dynamics for GS1 was carried out using satellite-retrieved soil moisture data, which revealed good agreement. The SWAT model (GS1) also captured the seasonal variability of the water balance components at the outlet of the watershed. This study demonstrated the potential to use remotely sensed evapotranspiration data for hydrological model calibration and validation in a sparsely gauged large river basin with reasonable accuracy. The novelty of the study is the use of these freely available satellite-derived AET datasets to effectively calibrate and validate an eco-hydrological model for a data-scarce catchment.

scholarly journals Assimilating scatterometer soil moisture data into conceptual hydrologic models at the regional scale

2006 ◽  
Vol 10 (3) ◽  
pp. 353-368 ◽  
Author(s):  
J. Parajka ◽  
V. Naeimi ◽  
G. Blöschl ◽  
W. Wagner ◽  
R. Merz ◽  
...  
Keyword(s):  
Soil Moisture ◽  
Regional Scale ◽  
Hydrologic Model ◽  
Hydrologic Models ◽  
Ungauged Catchments ◽  
Soil Moisture Dynamics ◽  
Model Efficiency ◽  
The Relationship ◽  
Runoff Model ◽  
Moisture Dynamics

Abstract. This paper examines the potential of scatterometer data from ERS satellites for improving hydrological simulations in both gauged and ungauged catchments. We compare the soil moisture dynamics simulated by a semidistributed hydrologic model in 320 Austrian catchments with the soil moisture dynamics inferred from the satellite data. The most apparent differences occur in the Alpine areas. Assimilating the scatterometer data into the hydrologic model during the calibration phase improves the relationship between the two soil moisture estimates without any significant decrease in runoff model efficiency. For the case of ungauged catchments, assimilating scatterometer data does not improve the daily runoff simulations but does provide more consistent soil moisture estimates. If the main interest is in obtaining estimates of catchment soil moisture, reconciling the two sources of soil moisture information seems to be of value because of the different error structures.

Change detection of storm runoff and sediment yield using hydrologic models following wildfire in a coastal redwood forest, California

2014 ◽  
Vol 44 (6) ◽  
pp. 572-581 ◽  
Author(s):  
Christopher G. Surfleet ◽  
Brian Dietterick ◽  
Arne Skaugset
Keyword(s):  
Sediment Load ◽  
Fire Severity ◽  
Uncertainty Assessment ◽  
Hydrologic Model ◽  
Stream Channels ◽  
Hydrologic Models ◽  
Alternative Analysis ◽  
Sediment Loads ◽  
Control Intervention ◽  
Paired Watershed

This study attempted to detect changes in stormflow volumes, peakflows, and sediment loads using hydrologic models within the context of an uncertainty assessment following wildfire. In 2009, after 8 years of study, the Lockheed Fire burned the treatment and control watersheds of a paired watershed study in coastal California, USA, eliminating the ability to continue a paired watershed before–after control–intervention (BACI) study design. An alternative analysis was used to detect stormflow and sediment load changes due to the wildfire by comparing measured posttreatment stormflow and sediment load with simulated predisturbance responses predicted with the hydrologic models HBV-EC and DHSVM. High natural variability of stormflow and sediment measurements compounded with uncertainty associated with the hydrologic models and climate suggest that only large changes can be detected. The fire and subsequent salvage harvest created an approximately 9%–12% reduction in forest overstory canopy and a 70%–90% consumption of understory vegetation. No discernible changes in slopes of regression lines were detected between predisturbance and postfire stormflow volumes, peakflows, or sediment loads. No changes were detected in stormflow volume, peakflow, or sediment loads comparing pre- and post-fire vegetation inputs to the hydrologic model DHSVM. The lack of detected change in streamflow to accelerate stream channel erosion combined with low to moderate fire severity adjacent to stream channels most likely were the reasons for no detected postfire change to sediment loads.

scholarly journals Spatial Pattern Oriented Multicriteria Sensitivity Analysis of a Distributed Hydrologic Model

Water ◽  
2018 ◽  
Vol 10 (9) ◽  
pp. 1188 ◽  
Author(s):  
Mehmet Demirel ◽  
Julian Koch ◽  
Gorka Mendiguren ◽  
Simon Stisen
Keyword(s):  
Sensitivity Analysis ◽  
Spatial Variability ◽  
Spatial Pattern ◽  
Spatial Patterns ◽  
Performance Measures ◽  
Sampling Strategy ◽  
Behavioral Model ◽  
Hydrologic Model ◽  
Actual Evapotranspiration ◽  
Model Parameters

Hydrologic models are conventionally constrained and evaluated using point measurements of streamflow, which represent an aggregated catchment measure. As a consequence of this single objective focus, model parametrization and model parameter sensitivity typically do not reflect other aspects of catchment behavior. Specifically for distributed models, the spatial pattern aspect is often overlooked. Our paper examines the utility of multiple performance measures in a spatial sensitivity analysis framework to determine the key parameters governing the spatial variability of predicted actual evapotranspiration (AET). The Latin hypercube one-at-a-time (LHS-OAT) sampling strategy with multiple initial parameter sets was applied using the mesoscale hydrologic model (mHM) and a total of 17 model parameters were identified as sensitive. The results indicate different parameter sensitivities for different performance measures focusing on temporal hydrograph dynamics and spatial variability of actual evapotranspiration. While spatial patterns were found to be sensitive to vegetation parameters, streamflow dynamics were sensitive to pedo-transfer function (PTF) parameters. Above all, our results show that behavioral model definitions based only on streamflow metrics in the generalized likelihood uncertainty estimation (GLUE) type methods require reformulation by incorporating spatial patterns into the definition of threshold values to reveal robust hydrologic behavior in the analysis.

scholarly journals Some Challenges in Hydrologic Model Calibration for Large-Scale Studies: A Case Study of SWAT Model Application to Mississippi-Atchafalaya River Basin

Hydrology ◽  
2019 ◽  
Vol 6 (1) ◽  
pp. 17 ◽  
Author(s):  
Narayanan Kannan ◽  
Chinnasamy Santhi ◽  
Michael J. White ◽  
Sushant Mehan ◽  
Jeffrey G. Arnold ◽  
...  
Keyword(s):  
United States ◽  
River Basin ◽  
Model Calibration ◽  
Large Scale ◽  
Swat Model ◽  
The United States ◽  
Hydrologic Model ◽  
Calibration Target ◽  
Atchafalaya River ◽  
Atchafalaya River Basin

This study is a part of the Conservation Effects Assessment Project (CEAP) aimed to quantify the environmental and economic benefits of conservation practices implemented in the cultivated cropland throughout the United States. The Soil and Water Assessment Tool (SWAT) model under the Hydrologic United Modeling of the United States (HUMUS) framework was used in the study. An automated flow calibration procedure was developed and used to calibrate runoff for each 8-digit watershed (within 20% of calibration target) and the partitioning of runoff into surface and sub-surface flow components (within 10% of calibration target). Streamflow was validated at selected gauging stations along major rivers within the river basin with a target R2 of >0.6 and Nash and Sutcliffe Efficiency of >0.5. The study area covered the entire Mississippi and Atchafalaya River Basin (MARB). Based on the results obtained, our analysis pointed out multiple challenges to calibration such as: (1) availability of good quality data, (2) accounting for multiple reservoirs within a sub-watershed, (3) inadequate accounting of elevation and slopes in mountainous regions, (4) poor representation of carrying capacity of channels, (5) inadequate capturing of the irrigation return flows, (6) inadequate representation of vegetative cover, and (7) poor representation of water abstractions (both surface and groundwater). Additional outstanding challenges to large-scale hydrologic model calibration were the coarse spatial scale of soils, land cover, and topography.

Sign in / Sign up

Export Citation Format

Share Document