Improving runoff prediction through the assimilation of the ASCAT soil moisture product

Abstract. The role and the importance of soil moisture for meteorological, agricultural and hydrological applications is widely known. Remote sensing offers the unique capability to monitor soil moisture over large areas (catchment scale) with, nowadays, a temporal resolution suitable for hydrological purposes. However, the accuracy of the remotely sensed soil moisture estimates has to be carefully checked. The validation of these estimates with in-situ measurements is not straightforward due the well-known problems related to the spatial mismatch and the measurement accuracy. The analysis of the effects deriving from assimilating remotely sensed soil moisture data into hydrological or meteorological models could represent a more valuable method to test their reliability. In particular, the assimilation of satellite-derived soil moisture estimates into rainfall-runoff models at different scales and over different regions represents an important scientific and operational issue. In this study, the soil wetness index (SWI) product derived from the Advanced SCATterometer (ASCAT) sensor onboard of the Metop satellite was tested. The SWI was firstly compared with the soil moisture temporal pattern derived from a continuous rainfall-runoff model (MISDc) to assess its relationship with modeled data. Then, by using a simple data assimilation technique, the linearly rescaled SWI that matches the range of variability of modelled data (denoted as SWI*) was assimilated into MISDc and the model performance on flood estimation was analyzed. Moreover, three synthetic experiments considering errors on rainfall, model parameters and initial soil wetness conditions were carried out. These experiments allowed to further investigate the SWI potential when uncertain conditions take place. The most significant flood events, which occurred in the period 2000–2009 on five subcatchments of the Upper Tiber River in Central Italy, ranging in extension between 100 and 650 km2, were used as case studies. Results reveal that the SWI derived from the ASCAT sensor can be conveniently adopted to improve runoff prediction in the study area, mainly if the initial soil wetness conditions are unknown.

Download Full-text

Improving runoff prediction through the assimilation of the ASCAT soil moisture product

Hydrology and Earth System Sciences ◽

10.5194/hess-14-1881-2010 ◽

2010 ◽

Vol 14 (10) ◽

pp. 1881-1893 ◽

Cited By ~ 254

Author(s):

L. Brocca ◽

F. Melone ◽

T. Moramarco ◽

W. Wagner ◽

V. Naeimi ◽

...

Keyword(s):

Soil Moisture ◽

Central Italy ◽

Model Performance ◽

Remotely Sensed ◽

Model Parameters ◽

Rainfall Runoff ◽

Catchment Scale ◽

Runoff Prediction ◽

Initial Soil ◽

Soil Wetness

Abstract. The role and the importance of soil moisture for meteorological, agricultural and hydrological applications is widely known. Remote sensing offers the unique capability to monitor soil moisture over large areas (catchment scale) with, nowadays, a temporal resolution suitable for hydrological purposes. However, the accuracy of the remotely sensed soil moisture estimates has to be carefully checked. The validation of these estimates with in-situ measurements is not straightforward due the well-known problems related to the spatial mismatch and the measurement accuracy. The analysis of the effects deriving from assimilating remotely sensed soil moisture data into hydrological or meteorological models could represent a more valuable method to test their reliability. In particular, the assimilation of satellite-derived soil moisture estimates into rainfall-runoff models at different scales and over different regions represents an important scientific and operational issue. In this study, the soil wetness index (SWI) product derived from the Advanced SCATterometer (ASCAT) sensor onboard of the Metop satellite was tested. The SWI was firstly compared with the soil moisture temporal pattern derived from a continuous rainfall-runoff model (MISDc) to assess its relationship with modeled data. Then, by using a simple data assimilation technique, the linearly rescaled SWI that matches the range of variability of modelled data (denoted as SWI*) was assimilated into MISDc and the model performance on flood estimation was analyzed. Moreover, three synthetic experiments considering errors on rainfall, model parameters and initial soil wetness conditions were carried out. These experiments allowed to further investigate the SWI potential when uncertain conditions take place. The most significant flood events, which occurred in the period 2000–2009 on five subcatchments of the Upper Tiber River in central Italy, ranging in extension between 100 and 650 km2, were used as case studies. Results reveal that the SWI derived from the ASCAT sensor can be conveniently adopted to improve runoff prediction in the study area, mainly if the initial soil wetness conditions are unknown.

Download Full-text

Identification of Hydrologic Models, Optimized Parameters, and Rainfall Inputs Consistent with In Situ Streamflow and Rainfall and Remotely Sensed Soil Moisture

Journal of Hydrometeorology ◽

10.1175/jhm-d-17-0240.1 ◽

2018 ◽

Vol 19 (8) ◽

pp. 1305-1320 ◽

Cited By ~ 6

Author(s):

Ashley J. Wright ◽

Jeffrey P. Walker ◽

Valentijn R. N. Pauwels

Keyword(s):

Time Series ◽

Soil Moisture ◽

Input Data ◽

Remotely Sensed ◽

Model Parameters ◽

Rainfall Time Series ◽

Hydrological Models ◽

Rainfall Runoff ◽

Model Input ◽

Rainfall Estimates

Abstract An increased understanding of the uncertainties present in rainfall time series can lead to improved confidence in both short- and long-term streamflow forecasts. This study presents an analysis that considers errors arising from model input data, model structure, model parameters, and model states with the objective of finding a self-consistent set that includes hydrological models, model parameters, streamflow, remotely sensed (RS) soil moisture (SM), and rainfall. This methodology can be used by hydrologists to aid model and satellite selection. Taking advantage of model input data reduction and model inversion techniques, this study uses a previously developed methodology to estimate areal rainfall time series for the study catchment of Warwick, Australia, for multiple rainfall–runoff models. RS SM observations from the Soil Moisture Ocean Salinity (SMOS) and Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) satellites were assimilated into three different rainfall–runoff models using an ensemble Kalman filter (EnKF). Innovations resulting from the observed and predicted SM were analyzed for Gaussianity. The findings demonstrate that consistency between hydrological models, model parameters, streamflow, RS SM, and rainfall can be found. Joint estimation of rainfall time series and model parameters consistently improved streamflow simulations. For all models rainfall estimates are less than the observed rainfall, and rainfall estimates obtained using the Sacramento Soil Moisture Accounting (SAC-SMA) model are the most consistent with gauge-based observations. The SAC-SMA model simulates streamflow that is most consistent with observations. EnKF innovations obtained when SMOS RS SM observations were assimilated into the SAC-SMA model demonstrate consistency between SM products.

Download Full-text

Event runoff calibration with LISEM in a recently burned Mediterranean forest catchment.

10.5194/egusphere-egu2020-11537 ◽

2020 ◽

Author(s):

Diana Vieira ◽

Marta Basso ◽

João Nunes ◽

Jacob Keizer ◽

Jantiene Baartman

Keyword(s):

Soil Moisture ◽

Moisture Content ◽

Field Data ◽

Soil Moisture Content ◽

Model Performance ◽

Hydrological Response ◽

Runoff Generation ◽

Catchment Scale ◽

Initial Soil ◽

Total Discharge

Wildfires are known to change post-fire hydrological response as a consequence of fire-induced changes such as soil water repellence (SWR). SWR has also been identified as a key factor determining runoff generation at plot and slope scale studies, in which soil moisture content (SMC) has been presented as dependent variable. However, these relationships have not been established at catchment scale yet, mainly due to the inherent difficulties in monitoring post-fire hydrological responses at this scale and in finding relationships between these events with SWR point (time and space) measurements. To fulfil these knowledge gaps, the present study aims to advance the knowledge on post-fire hydrological response by simulating quick flows from a small burned catchment using a physical event-based soil erosion model (OpenLISEM).OpenLISEM was applied to simulate sixteen events with two distinct initial soil moisture conditions (dry and wet), in which the model calibration was performed by adjusting Manning&#8217;s n and saturated soil moisture content (thetas). Considering that manual calibration resulted in distinct Manning&#8217;s n for wet and dry conditions, while thetas required an individual calibration for each event, an alternative parameterization of thetas was created by means of linear regressions, for all the events together (&#8220;overall&#8221;), and for wet and dry events separately (&#8220;wet&#8221; and &#8220;dry&#8221;). Model performance was evaluated at the outlet, while hillslope predictions were compared with runoff data from micro-plots that were installed at 3 of the hillslopes (Vieira et al., 2018).The validation of field data at micro-plot scale revealed several comparability limitations attributed to the time-step of the field data (1- to 2-weekly) in comparison to the duration of the events (170-940 min). Nevertheless, the most striking result from our simulations is the fact that OpenLISEM did not predict overland flow generation at two out of the three locations where it was observed. Our simulations also showed that the forest roads are a source of the runoff generation and their configuration affects catchment connectivity.At the outlet level, OpenLISEM achieved a satisfactory (0.50 < NSE &#8804; 0.70) and very good (NSE > 0.80) model performance according to Moriasi, et al. (2015), in predicting total discharge (NSE=0.95), peak discharge (NSE=0.68), and the time of the peak (NSE=1.00), for the entire set of events under manual calibration. In addition, simulations in wet conditions achieved higher accuracy in comparison to the dry ones.When using the parameterization based on the linear regression calibration, OpenLISEM simulation efficiency dropped, but still to satisfactory and very good (NSEoverall = 0.58, NSEcombined =0.86) accuracy levels for total discharge.Overall, we conclude that calibrating post-fire hydrological response at catchment scale with the OpenLISEM model, can result in reliable simulations for total flow, peak discharge and timing of the peaks. When considering the parameterization of thetas as proxy for repellent and wettable soils, more information than the initial soil moisture is required.

Download Full-text

SHETRAN and HEC HMS Model Evaluation for Runoff and Soil Moisture Simulation in the Jičinka River Catchment (Czech Republic)

Water ◽

10.3390/w13060872 ◽

2021 ◽

Vol 13 (6) ◽

pp. 872

Author(s):

Vesna Đukić ◽

Ranka Erić

Keyword(s):

Soil Moisture ◽

Czech Republic ◽

Hydrological Model ◽

Flash Flood ◽

Model Performance ◽

Rainfall Runoff ◽

The Czech Republic ◽

River Catchment ◽

Hydrological Variable ◽

Model Structures

Due to the improvement of computation power, in recent decades considerable progress has been made in the development of complex hydrological models. On the other hand, simple conceptual models have also been advanced. Previous studies on rainfall–runoff models have shown that model performance depends very much on the model structure. The purpose of this study is to determine whether the use of a complex hydrological model leads to more accurate results or not and to analyze whether some model structures are more efficient than others. Different configurations of the two models of different complexity, the Système Hydrologique Européen TRANsport (SHETRAN) and Hydrologic Modeling System (HEC-HMS), were compared and evaluated in simulating flash flood runoff for the small (75.9 km2) Jičinka River catchment in the Czech Republic. The two models were compared with respect to runoff simulations at the catchment outlet and soil moisture simulations within the catchment. The results indicate that the more complex SHETRAN model outperforms the simpler HEC HMS model in case of runoff, but not for soil moisture. It can be concluded that the models with higher complexity do not necessarily provide better model performance, and that the reliability of hydrological model simulations can vary depending on the hydrological variable under consideration.

Download Full-text

Climatic and basin factors affecting the flood frequency curve: PART II – A full sensitivity analysis based on the continuous simulation approach combined with a factorial experimental design

Hydrology and Earth System Sciences ◽

10.5194/hess-4-483-2000 ◽

2000 ◽

Vol 4 (3) ◽

pp. 483-498 ◽

Cited By ~ 6

Author(s):

M. Franchini ◽

A. M. Hashemi ◽

P. E. O’Connell

Keyword(s):

Sensitivity Analysis ◽

Soil Moisture ◽

Experimental Design ◽

Probability Distribution ◽

Flood Frequency ◽

Rainfall Regime ◽

Model Parameters ◽

Rainfall Runoff ◽

Factorial Experimental Design ◽

Frequency Curve

Abstract. The sensitivity analysis described in Hashemi et al. (2000) is based on one-at-a-time perturbations to the model parameters. This type of analysis cannot highlight the presence of parameter interactions which might indeed affect the characteristics of the flood frequency curve (ffc) even more than the individual parameters. For this reason, the effects of the parameters of the rainfall, rainfall runoff models and of the potential evapotranspiration demand on the ffc are investigated here through an analysis of the results obtained from a factorial experimental design, where all the parameters are allowed to vary simultaneously. This latter, more complex, analysis confirms the results obtained in Hashemi et al. (2000) thus making the conclusions drawn there of wider validity and not related strictly to the reference set selected. However, it is shown that two-factor interactions are present not only between different pairs of parameters of an individual model, but also between pairs of parameters of different models, such as rainfall and rainfall-runoff models, thus demonstrating the complex interaction between climate and basin characteristics affecting the ffc and in particular its curvature. Furthermore, the wider range of climatic regime behaviour produced within the factorial experimental design shows that the probability distribution of soil moisture content at the storm arrival time is no longer sufficient to explain the link between the perturbations to the parameters and their effects on the ffc, as was suggested in Hashemi et al. (2000). Other factors have to be considered, such as the probability distribution of the soil moisture capacity, and the rainfall regime, expressed through the annual maximum rainfalls over different durations. Keywords: Monte Carlo simulation; factorial experimental design; analysis of variance (ANOVA)

Download Full-text

Sensitivity of the PBL and Precipitation in 12-Day Simulations of Warm-Season Convection Using Different Land Surface Models and Soil Wetness Conditions

Monthly Weather Review ◽

10.1175/2007mwr2289.1 ◽

2008 ◽

Vol 136 (7) ◽

pp. 2321-2343 ◽

Cited By ~ 52

Author(s):

S. B. Trier ◽

F. Chen ◽

K. W. Manning ◽

M. A. LeMone ◽

C. A. Davis

Keyword(s):

United States ◽

Soil Moisture ◽

Land Surface ◽

Warm Season ◽

Land Surface Models ◽

Surface Conditions ◽

Late Afternoon ◽

Surface Models ◽

Initial Soil ◽

Soil Wetness

Abstract A coupled land surface–atmospheric model that permits grid-resolved deep convection is used to examine linkages between land surface conditions, the planetary boundary layer (PBL), and precipitation during a 12-day warm-season period over the central United States. The period of study (9–21 June 2002) coincided with an extensive dry soil moisture anomaly over the western United States and adjacent high plains and wetter-than-normal soil conditions over parts of the Midwest. A range of possible atmospheric responses to soil wetness is diagnosed from a set of simulations that use land surface models (LSMs) of varying sophistication and initial land surface conditions of varying resolution and specificity to the period of study. Results suggest that the choice of LSM [Noah or the less sophisticated simple slab soil model (SLAB)] significantly influences the diurnal cycle of near-surface potential temperature and water vapor mixing ratio. The initial soil wetness also has a major impact on these thermodynamic variables, particularly during and immediately following the most intense phase of daytime surface heating. The soil wetness influences the daytime PBL evolution through both local and upstream surface evaporation and sensible heat fluxes, and through differences in the mesoscale vertical circulation that develops in response to horizontal gradients of the latter. Resulting differences in late afternoon PBL moist static energy and stability near the PBL top are associated with differences in subsequent late afternoon and evening precipitation in locations where the initial soil wetness differs among simulations. In contrast to the initial soil wetness, soil moisture evolution has negligible effects on the mean regional-scale thermodynamic conditions and precipitation during the 12-day period.

Download Full-text

An Experimental Study on the Rainfall-Runoff Progress of Wheat under Different Slope Angle

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.912-914.1986 ◽

2014 ◽

Vol 912-914 ◽

pp. 1986-1994

Author(s):

Na Na Zhao ◽

Fu Liang Yu ◽

Chuan Zhe Li ◽

Jia Liu ◽

Hao Wang

Keyword(s):

Regression Analysis ◽

Soil Moisture ◽

Power Function ◽

Surface Runoff ◽

Rainfall Intensity ◽

Slope Angle ◽

Subsurface Flow ◽

Rainfall Runoff ◽

Determination Coefficient ◽

Initial Soil

Rainfall-runoff process plays an important role in hydrological cycle, and the study on the rainfall-runoff will provide foundation and basis for research on basin hydrology and flood forecasting. In this paper, the surface runoff and subsurface flow of wheat were observed in the laboratory by artificial rainfall, and analyzed the cumulated surface runoff and recession process of subsurface flow by regression analysis. In addition, the factors affected the runoff and response of soil moisture on the runoff coefficients was also discussed. Results showed that the rainfall intensity, soil coverage and slope had important influence on the surface runoff generation, and the surface runoff was observed when the total rainfall amount exceeded 32mm and 13mm for 5°and 15° slope respectively. The cumulative surface runoff could be expressed as a power function, which had higher determination coefficient R2 (0.92~0.999). The subsurface flow was only observed at the ripening period and wheat stubble treatment, and mainly affected by slope angle and initial soil moisture, whereas rainfall intensity showed little impact. The recession curve of subsurface flow can be described as a simple exponential expression or power function, which the determination coefficient was 0.88 and 0.94 by regression analysis, respectively. Moreover, there was an obvious threshold (approximately 30%) between the average initial soil moisture and runoff coefficients, which the runoff increased significantly as above the threshold.

Download Full-text

An experimental approach to reduce the parametric dimensionality for rainfall–runoff models

Hydrology Research ◽

10.2166/nh.2016.145 ◽

2016 ◽

Vol 48 (1) ◽

pp. 48-65 ◽

Cited By ~ 2

Author(s):

Jie Chen ◽

Richard Arsenault ◽

François P. Brissette

Keyword(s):

Experimental Approach ◽

Model Performance ◽

Optimal Number ◽

Model Parameters ◽

Hydrological Models ◽

Rainfall Runoff ◽

Efficiency Criterion ◽

Performance Loss ◽

The Past ◽

Multi Objective Genetic Algorithm

Sobol’ sensitivity analysis has been used successfully in the past to reduce the parametric dimensionality for hydrological models. However, the effects of its limitation, in that it assumes an independence of parameters, need to be investigated. This study proposes an experimental approach to assess the commonly used Sobol’ analysis for reducing the parameter dimensionality of hydrological models. In this approach, the number of model parameters is directly pitted against an efficiency criterion within a multi-objective genetic algorithm (MOGA), thus allowing both the identification of key model parameters and the optimal number of parameters to be used within the same analysis. The proposed approach was tested and compared with the Sobol’ method based on a conceptual lumped hydrological model (HSAMI) with 23 free parameters. The results show that both methods performed very similarly, and allowed 11 out of 23 HSAMI parameters to be reduced with little loss in model performance. Based on this comparison, Sobol’ appears to be an effective and robust method despite its limitations. On the other hand, the MOGA algorithm outperformed Sobol’ analysis for further reduction of the parametric space and found optimal solutions with as few as eight parameters with minimal performance loss in validation.

Download Full-text

Added-value of satellite soil moisture assimilation in hydrological modelling: an evaluation through a large experiment over Europe

10.5194/egusphere-egu2020-16239 ◽

2020 ◽

Author(s):

Domenico De Santis ◽

Christian Massari ◽

Stefania Camici ◽

Sara Modanesi ◽

Luca Brocca ◽

...

Keyword(s):

Surface Layer ◽

Soil Moisture ◽

Winter Season ◽

Assimilation Efficiency ◽

Remotely Sensed ◽

Open Loop ◽

Added Value ◽

Observation Error ◽

Loop Model ◽

Rainfall Runoff

The increasing availability of remotely sensed soil moisture (SM) observations has brought great interest in their use in data assimilation (DA) frameworks in order to improve streamflow simulations. However, the added-value of assimilating satellite SM into rainfall-runoff models is still difficult to be quantified, and much more research is needed to fully understand benefits and limitations.Here, an extensive evaluation of remotely sensed SM assimilation on hydrological model performances was carried out, involving 775 catchments across Europe. Satellite observations for over a decade from the three ESA CCI SM products (ACTIVE, PASSIVE and COMBINED) were assimilated in a lumped rainfall-runoff model which includes a thin surface layer in its soil schematization, by using the Ensemble Kalman Filter (EnKF). Observations were mapped into the space of modelled surface layer SM through a monthly CDF-matching prior to DA, while the observation error variance was calibrated in every catchment in order to maximize the assimilation efficiency.The implemented DA procedure, aimed at reducing only random errors in SM variables, generally resulted in limited runoff improvements, although with some variability within the study domain. Factors emerging as relevant for the assessment of assimilation impact were: i) the open-loop (OL) model performance; ii) the remotely sensed SM accuracy for hydrological purposes; iii) the sensitivity of the catchment response to soil moisture dynamics; and also iv) issues in DA implementation (e.g., violations in theoretical assumptions).The open-loop model results contributed significantly to explain differences in assimilation performances observed within the study area as well as at the seasonal scale; overall, the high OL efficiency is the main cause of the slight improvements here observed after DA. The integration of satellite SM information, showing greater skills in correspondence of poorer streamflow simulations, confirmed a potential in reducing the effects of rainfall inaccuracies.The variability in satellite SM accuracy for hydrological purposes was also found to be relevant in DA assessment. The ACTIVE product assimilation generally provided the best streamflow results within the study catchments, followed by COMBINED and PASSIVE ones, while factors affecting the SM retrieval such as vegetation density and topographic complexity were not found to have a decisive effect on DA results.Low assimilation performances were obtained when runoff was dominated by snow dynamics (e.g., in the northern areas of the study domain, or in winter season at medium latitudes), due to the SM conditions having a negligible effect on the hydrological response.Finally, in basins where SM was persistently near the saturation value, deteriorations in hydrological simulations were observed, mainly attributable to violation of error normality hypothesis in EnKF due to the bounded nature of soil moisture.In conclusion, the added-value of assimilating remotely sensed SM into rainfall-runoff models was confirmed to be linked to multiple factors: understanding their contribution and interactions deserves further research and is fundamental to take full advantage of the potential of satellite SM retrievals, in parallel with their progress in terms of accuracy and resolutions.

Download Full-text