Modelling the water balance of a mesoscale catchment basin using remotely sensed land cover data

Hydrological modelling of mesoscale catchments is often adversely affected by a lack of adequate information about specific site conditions. In particular, digital land cover data are available from data sets which were acquired on a European or a national scale. These data sets do not only exhibit a restricted spatial resolution but also a differentiation of crops and impervious areas which is not appropriate to the needs of mesoscale hydrological models. In this paper, the impact of remote sensing data on the reliability of a water balance model is investigated and compared to model results determined on the basis of CORINE (Coordination of Information on the Environment) Land Cover as a reference. The aim is to quantify the improved model performance achieved by an enhanced land cover representation and corresponding model modifications. Making use of medium resolution satellite imagery from SPOT, LANDSAT ETM+ and ASTER, detailed information on land cover, especially agricultural crops and impervious surfaces, was extracted over a 5-year period (2000–2004). Crop-specific evapotranspiration coefficients were derived by using remote sensing data to replace grass reference evapotranspiration necessitated by the use of CORINE land cover for rural areas. For regions classified as settlement or industrial areas, degrees of imperviousness were derived. The data were incorporated into the hydrological model GROWA (large-scale water balance model), which uses an empirical approach combining distributed meteorological data with distributed site parameters to calculate the annual runoff components. Using satellite imagery in combination with runoff data from gauging stations for the years 2000–2004, the actual evapotranspiration calculation in GROWA was methodologically extended by including empirical crop coefficients for actual evapotranspiration calculations. While GROWA originally treated agricultural areas as homogeneous, now a consideration and differentiation of the main crops is possible. The accuracy was determined by runoff measurements from gauging stations. Differences in water balances resulting from the use of remote sensing data as opposed to CORINE were analysed in this study using a representative subcatchment. Resulting Nash–Sutcliff model efficiencies improved from 0.372 to 0.775 and indicate that the enhanced model can produce thematically more accurate and spatially more detailed local water balances. However, the proposed model enhancements by satellite imagery have not exhausted the full potential of water balance modelling, for which a higher temporal resolution is required.     

The use of remotely sensed land cover to derive floodplain friction coefficients for flood inundation modelling

Remotely sensed land cover was used to generate spatially‐distributed friction coefficients for use in a two‐dimensional model of flood inundation. Such models are at the forefront of research into the prediction of river flooding. Standard practice, however, is to use single (static) friction coefficients on both the channel and floodplain, which are varied in a calibration procedure to provide a “best fit” to a known inundation extent. Spatially‐distributed friction provides a physically grounded estimate of friction that does not require fitting to a known inundation extent, but which can be fitted if desired. Remote sensing offers the opportunity to map these friction coefficients relatively straightforwardly and for low cost. Inundation was predicted using the LISFLOOD‐FP model for a reach on the River Nene, UK. Friction coefficients were produced from land cover predicted from Landsat TM imagery using both ML and fuzzy c‐means classifiction. The elevetion data used were from combined contour and differential global positioning system (GPS) elevation data. Predicted inundation using spatially‐distributed and static friction were compared. Spatially‐distributed friction had the greatest effect on the timing of flood inundation, but a small effect on predicted inundation extent. The results indicate that spatially‐distributed friction should be considered where the timing of initial flooding (e.g. for early warning) is important. Copyright © 2007 John Wiley & Sons, Ltd.     

Modelling rating curves using remotely sensed LiDAR data

Accurate stream discharge measurements are important for many hydrological studies. In remote locations, however, it is often difficult to obtain stream flow information because of the difficulty in making the discharge measurements necessary to define stage‐discharge relationships (rating curves). This study investigates the feasibility of defining rating curves by using a fluid mechanics‐based model constrained with topographic data from an airborne LiDAR scanning. The study was carried out for an 8m‐wide channel in the boreal landscape of northern Sweden. LiDAR data were used to define channel geometry above a low flow water surface along the 90‐m surveyed reach. The channel topography below the water surface was estimated using the simple assumption of a flat streambed. The roughness for the modelled reach was back calculated from a single measurment of discharge. The topographic and roughness information was then used to model a rating curve. To isolate the potential influence of the flat bed assumption, a ‘hybrid model’ rating curve was developed on the basis of data combined from the LiDAR scan and a detailed ground survey. Whereas this hybrid model rating curve was in agreement with the direct measurements of discharge, the LiDAR model rating curve was equally in agreement with the medium and high flow measurements based on confidence intervals calculated from the direct measurements. The discrepancy between the LiDAR model rating curve and the low flow measurements was likely due to reduced roughness associated with unresolved submerged bed topography. Scanning during periods of low flow can help minimize this deficiency. These results suggest that combined ground surveys and LiDAR scans or multifrequency LiDAR scans that see ‘below’ the water surface (bathymetric LiDAR) could be useful in generating data needed to run such a fluid mechanics‐based model. This opens a realm of possibility to remotely sense and monitor stream flows in channels in remote locations. Copyright © 2012 John Wiley & Sons, Ltd.     

An information-based criterion to measure pixel-level thematic uncertainty in land cover classifications

Traditional accuracy assessment of satellite-derived maps relies on a confusion matrix and its associated indices built by comparing ground truth observations and classification outputs at specific locations. These indices may be applied at the map-level or at the class level. However, the spatial variation of the accuracy is not captured by those statistics. Pixel-level thematic uncertainty measures derived from class membership probability vectors can provide such spatially explicit information. In this paper, a new information-based criterion—the equivalent reference probability—is introduced to provide a synoptic thematic uncertainty measure that has the advantage of taking the maximum probability value into account while committing for the full set of probabilities. The fundamental theoretical properties of this indicator was first highlighted and its use was afterwards demonstrated on a real case study in Belgium. Results showed that the proposed approach positively correlates with the quality of the classification and is more sensitive than the classical maximum probability criterion. As this information-based criterion can be used for providing spatially explicit maps of thematic uncertainty quality, it provides substantial additional information regarding classification quality compared to conventional quality measures. Accordingly, it proved to be useful both for end-users and map producers as a way to better understand the nature of the errors and to subsequently improve the map quality.     

Physical statistical algorithm for precipitable water vapor inversion on land surface based on multi-source remotely sensed data

Water vapor plays a crucial role in atmospheric processes that act over a wide range of temporal and spatial scales, from global climate to micrometeorology. The determination of water vapor distribution in the atmosphere and its changing pattern is very important. Although atmospheric scientists have developed a variety of means to measure precipitable water vapor(PWV) using remote sensing data that have been widely used, there are some limitations in using one kind satellite measurements for PWV retrieval over land. In this paper, a new algorithm is proposed for retrieving PWV over land by combining different kinds of remote sensing data and it would work well under the cloud weather conditions. The PWV retrieval algorithm based on near infrared data is more suitable to clear sky conditions with high precision. The 23.5 GHz microwave remote sensing data is sensitive to water vapor and powerful in cloud-covered areas because of its longer wavelengths that permit viewing into and through the atmosphere. Therefore, the PWV retrieval results from near infrared data and the indices combined by microwave bands remote sensing data which are sensitive to water vapor will be regressed to generate the equation for PWV retrieval under cloud covered areas. The algorithm developed in this paper has the potential to detect PWV under all weather conditions and makes an excellent complement to PWV retrieved by near infrared data. Different types of surface exert different depolarization effects on surface emissions, which would increase the complexity of the algorithm. In this paper, MODIS surface classification data was used to consider this influence. Compared with the GPS results, the root mean square error of our algorithm is 8 mm for cloud covered area. Regional consistency was found between the results from MODIS and our algorithm. Our algorithm can yield reasonable results on the surfaces covered by cloud where MODIS cannot be used to retrieve PWV.     

Intercomparison of the JULES and CABLE land surface models through assimilation of remotely sensed soil moisture in southeast Australia

Numerous land surface models exist for predicting water and energy fluxes in the terrestrial environment. These land surface models have different conceptualizations (i.e., process or physics based), together with structural differences in representing spatial variability, alternate empirical methods, mathematical formulations and computational approach. These inherent differences in modeling approach, and associated variations in outputs make it difficult to compare and contrast land surface models in a straight-forward manner. While model intercomparison studies have been undertaken in the past, leading to significant progress on the improvement of land surface models, additional framework towards identification of model weakness is needed. Given that land surface models are increasingly being integrated with satellite based estimates to improve their prediction skill, it is practical to undertake model intercomparison on the basis of soil moisture data assimilation. Consequently, this study compares two land surface models: the Joint UK Land Environment Simulator (JULES) and the Community Atmosphere Biosphere Land Exchange (CABLE) for soil moisture estimation and associated assessment of model uncertainty. A retrieved soil moisture data set from the Soil Moisture and Ocean Salinity (SMOS) mission was assimilated into both models, with their updated estimates validated against in-situ soil moisture in the Yanco area, Australia. The findings show that the updated estimates from both models generally provided a more accurate estimate of soil moisture than the open loop estimate based on calibration alone. Moreover, the JULES output was found to provide a slightly better estimate of soil moisture than the CABLE output at both near-surface and deeper soil layers. An assessment of the updated membership in decision space also showed that the JULES model had a relatively stable, less sensitive, and more highly convergent internal dynamics than the CABLE model.     

Novel fuzzy uncertainty modeling for land cover classification based on clustering analysis

It is well known that there is a degree of fuzzy uncertainty in land cover classification using remote sensing (RS) images. In this article, we propose a novel fuzzy uncertainty modeling algorithm for representing the features of land cover patterns, and present an adaptive interval type-2 fuzzy clustering method. The proposed fuzzy uncertainty modeling method is performed in two main phases. First, the segmentation units of the input multi-spectral RS image data are subjected to objectbased interval-valued symbolic modeling. As a result, features for each land cover type are represented in the form of an intervalvalued symbolic vector, which describes the intra-class uncertainty better than the source data and improves the separability between different classes. Second, interval type-2 fuzzy sets are generated for each cluster based on the distance metric of the interval-valued vectors. This step characterizes the inter-class high-order fuzzy uncertainty and improves the classification accuracy. To demonstrate the advantages and effectiveness of the proposed approach, extensive experiments are conducted on two multispectral RS image datasets from regions with complex land cover characteristics, and the results are compared with those given by well-known fuzzy and conventional clustering algorithms.     

Investigation of the snow‐cover dynamics in the Upper Euphrates Basin of Turkey using remotely sensed snow‐cover products and hydrometeorological data

The Euphrates and Tigris rivers serve as the most important water resources in the Middle East. Precipitation in this region falls mostly in the form of snow over the higher elevations of the Euphrates Basin and remains on the ground for nearly half of the year. This snow‐covered area (SCA) is a key element of the hydrological cycle, and monitoring the SCA is crucial for making accurate forecasts of snowmelt discharge, especially for energy production, flood control, irrigation, and reservoir‐operation optimization in the Upper Euphrates (Karasu) Basin. Remote sensing allows the detection of the spatio‐temporal patterns of snow cover across large areas in inaccessible terrain, such as the eastern part of Turkey, which is highly mountainous. In this study, a seasonal evaluation of the snow cover from 2000 to 2009 was performed using 8‐day snow‐cover products (MOD10C2) and the daily snow‐water equivalent (SWE) product. The values of SWE products were obtained using an assimilation process based on the Helsinki University of Technology model using equal area Special Sensor Microwave Imager (SSM/I) Earth‐gridded advanced microwave scanning radiometer—EOS daily brightness‐temperature values. In the Karasu Basin, the SCA percentage for the winter period is 80–90%. The relationship between the SCA and the runoff during the spring period is analysed for the period from 2004 to 2009. An inverse linear relationship between the normalized SCA and the normalized runoff values was obtained (r = 0·74). On the basis of the monthly mean temperature, total precipitation and snow depth observed at meteorological stations in the basin, the decrease in the peak discharges, and early occurrences of the peak discharges in 2008 and 2009 are due to the increase in the mean temperature and the decrease in the precipitation in April. Copyright © 2011 John Wiley & Sons, Ltd.     

Quantifying the relative impact of hydrological and hydraulic modelling parameterizations on uncertainty of inundation maps

ABSTRACT

Flood risk management strongly relies on inundation models for river basin zoning in flood-prone and risk-free areas. Floodplain zoning is significantly affected by the diverse and concurrent uncertainties that characterize the modelling chain used for producing inundation maps. In order to quantify the relative impact of the uncertainties linked to a lumped hydrological (rainfall–runoff) model and a FLO-2D hydraulic model, a Monte Carlo procedure is proposed in this work. The hydrological uncertainty is associated with the design rainfall estimation method, while the hydraulic model uncertainty is associated with roughness parameterization. This uncertainty analysis is tested on the case study of the Marta coastal catchment in Italy, by comparing the different frequency, extent and depth of inundation simulations associated with varying rainfall forcing and/or hydraulic model roughness realizations. The results suggest a significant predominance of the hydrological uncertainty with respect to the hydraulic one on the overall uncertainty associated with the simulated inundation maps.     

A ground validation problem of remotely sensed soil moisture data

 As the use of space-based sensors to observe soil moisture is becoming more plausible, it is becoming necessary to validate the remotely sensed soil moisture retrieval algorithms. In this paper, measurements of point gauges on the ground are analyzed as a possible ground-truth source for the comparison with remotely sensed data. The design compares a sequence of measurements taken on the ground and from space. The authors review the mean square error of expected differences between the two systems by Ha and North (1994), which is applied to the Little Washita watershed using the soil moisture dynamics model developed by Entekhabi and Rodriguez-Iturbe (1994). The model parameters estimated by Yoo and Shin (1998) for the Washita `92 (relative) soil moisture data are used in this study. By considering about 20 pairs of ground- and space-based measure-ments (especially, for the same case as the Washita `92 that the space-based sensor visits the FOV once a day), the expected error was able to be reduced to approximately 10 of the standard deviation of the fluctuations of the system alone. This seems to be an acceptable level of tolerance for identifying biases in the retrieval algorithms.     

Assimilating remotely sensed snow observations into a macroscale hydrology model

Accurate forecasting of snow properties is important for effective water resources management, especially in mountainous areas like the western United States. Current model-based forecasting approaches are limited by model biases and input data uncertainties. Remote sensing offers an opportunity for observation of snow properties, like areal extent and water equivalent, over larger areas. Data assimilation provides a framework for optimally merging information from remotely sensed observations and hydrologic model predictions. An ensemble Kalman filter (EnKF) was used to assimilate remotely sensed snow observations into the variable infiltration capacity (VIC) macroscale hydrologic model over the Snake River basin. The snow cover extent (SCE) product from the moderate resolution imaging spectroradiometer (MODIS) flown on the NASA Terra satellite was used to update VIC snow water equivalent (SWE), for a period of four consecutive winters (1999–2003). A simple snow depletion curve model was used for the necessary SWE–SCE inversion. The results showed that the EnKF is an effective and operationally feasible solution; the filter successfully updated model SCE predictions to better agree with the MODIS observations and ground surface measurements. Comparisons of the VIC SWE estimates following updating with surface SWE observations (from the NRCS SNOTEL network) indicated that the filter performance was a modest improvement over the open-loop (un-updated) simulations. This improvement was more evident for lower to middle elevations, and during snowmelt, while during accumulation the filter and open-loop estimates were very close on average. Subsequently, a preliminary assessment of the potential for assimilating the SWE product from the advanced microwave scanning radiometer (AMSR-E, flown on board the NASA Aqua satellite) was conducted. The results were not encouraging, and appeared to reflect large errors in the AMSR-E SWE product, which were also apparent in comparisons with SNOTEL data.     

Patterns of remotely sensed floodplain saturation and its use in runoff predictions

Principal components analysis (PCA) is applied to a time series of European Remote Sensing (ERS) synthetic aperture radar (SAR) scenes of the Alzette River floodplain (Grand‐Duchy of Luxembourg). These images cover markedly different hydrological conditions during several winter seasons in order to enable the examination of the decrease of the radar backscattering signal during drying‐up phases following important flood events. At the floodplain scale, with homogeneous land use and constant topography, the first principal components (PCs) are mainly dominated by the variance related to the changing areas. The PCs are thus mainly controlled by subsurface and surface water dynamics. The field observations of a densely equipped piezometric network in the floodplain are used to calculate a mean soil saturation index (SSI) continuously. A classification scheme, based on the PCs and k‐means algorithm, leads to the segmentation of the floodplain into several hydrological behaviour classes with distinctive responses versus changing moisture conditions. To validate this classification method with ground‐based estimations, the relation between the mean backscattering values of microplots within each PCA‐derived hydrological class and the water table measurements, expressed by means of the SSI, is evaluated. Results show that each class of microplots is characterized by the slope of the ‘backscattering–SSI’ function and by the SSI threshold value at which groundwater resurgence appears. The water ponding implies very low signal return due to the specular backscattering effect on the water surface. Based on established relationships between measured initial water table depths, runoff coefficients and rainfall‐induced water table rises, these results are used to discuss the potential of SAR‐derived information in flood management applications. Copyright © 2005 John Wiley & Sons, Ltd.     

Monitoring turbid freshwater plume characteristics by means of remotely sensed data

Remotely sensed data may provide easy access for monitoring the spatial separation and obtaining the hydrodynamic characteristics of turbid freshwater plumes created by river flow in the marine environment. Traditional methods are time consuming and require great effort to produce sufficient data. In this project, integrated research has been carried out on a river to demonstrate the utility of remote sensing (RS) technology for studying fundamental theoretical properties of turbulent mixtures. The Filyos River mouth, located on the Black Sea coast of Turkey, is the research area. Flow properties, such as the horizontal dispersion coefficient, have been calculated (using Landsat TM sensor images taken on two different dates). The effects of the plume on the morphology of neighbouring beaches are also examined. This study shows the utility of RS technology for generating quantitative data and better defining the hydraulic behaviour of a river with high turbidity. Copyright © 2006 John Wiley & Sons, Ltd.     

Recharge estimation using remotely sensed evapotranspiration in an irrigated catchment in southeast Australia

Reliable estimates of groundwater recharge are required for the sustainable management of surface and ground water resources in semi‐arid regions particularly in irrigated regions. In this study, groundwater recharge was estimated for an irrigated catchment in southeast Australia using a semi‐distributed hydrological model (SWAT). The model was calibrated under the dry climatic conditions for the period from August 2002 to July 2003 using flow and remotely sensed evapotranspiration (ET). The model was able to simulate observed monthly drain flow and spatially distributed remotely sensed ET. Recharge tended to be higher for irrigated land covers, such as perennial pasture, than for non‐irrigated land. On average, the estimated annual catchment recharge ranged between 147 and 289 mm which represented about 40% of the total rainfall and irrigation inputs. The optimized soil parameters indirectly reflected flow bypassing the soil matrix that could be responsible for this substantial amount of recharge. Overall, the estimated recharge was much more than that previously estimated for the wetter years. Copyright © 2011 John Wiley & Sons, Ltd.     

Identifying baseflow source areas using remotely sensed and ground-based hydrologic data

Understanding how rainfall and snowmelt influence baseflow, the groundwater-fed component of streamflow, is essential for sound water resources management. Current approaches to understand the spatial couplings between these processes and baseflow are limited. The most commonly used methods include geochemical tracers and hydrologic models. A key limitation of the first is cost, while the second is limited by the need for simplifying assumptions. This study developed a data-driven approach which leverages satellite Earth observation data and ground-based data to assess the degree to which baseflow is influenced by upstream rainfall and snowmelt in California's Sierra Nevada. The procedure involved: (1) separation of baseflow from streamflow time series using a low-pass filtering technique, (2) quantification of aquifer drainage timescales through baseflow recession analysis, (3) application of time series and information theory methods to identify the areas which have the greatest impacts on baseflow through both rainfall and snowmelt, and (4) characterization of the elevation zones which have a prevailing influence on baseflow. Results suggest that areas which have the strongest impact on baseflow through rainfall and snowmelt are not necessarily the areas which experience the highest annual rates of snowmelt or rainfall; snowmelt occurring in the 3000–3700 m elevation range was found to be the most important driver of baseflow.     

Improvement of dark object method in atmospheric correction of hyperspectral remotely sensed data

The existing methods in atmospheric correction of hyperspectral data usually focus on removing the effects of water vapor and other absorptive gases, while this paper mainly studies the method of re- moving the influence of the aerosol and the water vapor simultaneously. Because the hyperspectral data has a larger number of bands, the conventional dark object method cannot be applied to the at- mospheric correction of the hyperspectral data which can be improved, as described in this paper, by adequately making use of spectral characteristics of the hyperspectral data with an iterative correction during the whole process. The effects of the aerosol and water vapor are eliminated at the same time finally. The improved dark object method is used to do the atomospheric correction of the Hyperion data in Yanzhou, Shandong Province as an example. And the result indicates that it can correct the atmospheric influence of the hyperspectral data quickly and remarkably.     

Conditional simulation of remotely sensed rainfall data using a non-Gaussian v-transformed copula

Quantification of rainfall and its spatial and temporal variability is extremely important for reliable hydrological and meteorological modeling. While rain gauge measurements do not provide reasonable areal representation of rainfall, remotely sensed precipitation estimates offer much higher spatial resolution. However, uncertainties associated with remotely sensed rainfall estimates are not well quantified. This issue is important considering the fact that uncertainties in input rainfall are the main sources of error in hydrologic processes. Using an ensemble of rainfall estimates that resembles multiple realizations of possible true rainfall, one can assess uncertainties associated with remotely sensed rainfall data. In this paper, ensembles are generated by imposing rainfall error fields over remotely sensed rainfall estimates. A non-Gaussian copula-based model is introduced for simulation of rainfall error fields. The v-transformed copula is employed to describe the dependence structure of rainfall error estimates without the influence of the marginal distribution. Simulations using this model can be performed unconditionally or conditioned on ground reference measurements such that rain gauge data are honored at their locations. The presented model is implemented for simulation of rainfall ensembles across the Little Washita watershed, Oklahoma. The results indicate that the model generates rainfall fields with similar spatio-temporal characteristics and stochastic properties to those of observed rainfall data.     

Suitability mapping of global wetland areas and validation with remotely sensed data

With increasing urbanization and agricultural expansion, large tracts of wetlands have been either disturbed or converted to other uses. To protect wetlands, accurate distribution maps are needed. However, because of the dramatic diversity of wetlands and difficulties in field work, wetland mapping on a large spatial scale is very difficult to do. Until recently there were only a few high resolution global wetland distribution datasets developed for wetland protection and restoration. In this paper, we used hydrologic and climatic variables in combination with Compound Topographic Index(CTI) data in modeling the average annual water table depth at 30 arc-second grids over the continental areas of the world except for Antarctica. The water table depth data were modeled without considering influences of anthropogenic activities. We adopted a relationship between potential wetland distribution and water table depth to develop the global wetland suitability distribution dataset. The modeling results showed that the total area of global wetland reached 3.316×107 km2. Remote-sensing-based validation based on a compilation of wetland areas from multiple sources indicates that the overall accuracy of our product is 83.7%. This result can be used as the basis for mapping the actual global wetland distribution. Because the modeling process did not account for the impact of anthropogenic water management such as irrigation and reservoir construction over suitable wetland areas, our result represents the upper bound of wetland areas when compared with some other global wetland datasets. Our method requires relatively fewer datasets and has a higher accuracy than a recently developed global wetland dataset.     

Integration of remotely sensed C factor into SWAT for modelling sediment yield

The C factor, representing the impact of plant and ground cover on soil loss, is one of the important factors of the Modified Universal Soil Loss Equation (MUSLE) in the Soil and Water Assessment Tool (SWAT) to model sediment yield. The daily update of C factors in SWAT was originally determined by land use types and plant growth cycles. This does not reflect the spatial variation of C values that exists within a large land use area. We present a new approach to integrate remotely sensed C factors into SWAT for highlighting the effect of detailed vegetative cover data on soil erosion and sediment yield. First, the C factor was estimated using the abundance of ground components extracted from remote sensing images. Then, the gridding data of the C factor were aggregated to hydrological response units (HRUs), instead of to land use units of SWAT. In the end, the C factor values in HRUs were integrated into SWAT to predict sediment yield by modifying the ysed subroutine. This substitution work not only increases the spatial variation of the C factor in SWAT, but also makes it possible to utilize other sources of C databases rather than those from the United States. The demonstration in the Dage basin shows that the modified SWAT produces reasonable results in water flow simulation and sediment yield prediction using remotely sensed C values. The Nash–Sutcliffe efficiency coefficient (E_NS) and R² for surface runoff range from 0·69 to 0·77 and 0·73 to 0·87, respectively. The coefficients E_NS and R² for sediment yield were generally above 0·70 and 0·60, respectively. The soil erosion risk map based on sediment yield prediction at the HRU level illustrates instructive details on spatial distribution of soil loss. Copyright © 2011 John Wiley & Sons, Ltd.