Search

• Water-level based rainfall-runoff calibration can be improved using discharge index. • Efficacy of regionalized discharge index is comparable with that of observed values. • Water-level based calibration with discharge index works better for wet catchments. • Regionalized Qmean is a better choice than regionalized Q95 as discharge index. Rainfall-runoff models are generally calibrated by using continuous stream discharge data. However, most catchments around the globe remain ungauged due to the difficulty in installing gauges in river channels with complex morphology and due to the high costs of installation and maintenance. Recently, new calibration methods that use water level data instead of gauged discharge to calibrate rainfall-runoff models have been proposed. With emerging low-cost local water-level monitoring technologies or with satellite altimetry, these water-level based calibration methods have the potential to significantly extend runoff prediction capability. However, rainfall-runoff models calibrated from using water-level information alone may lead to biased discharge even when water-level variations are accurately reproduced. It has been shown that this bias problem can be significantly alleviated when just a few discharge measurements, especially of high flows, are available and incorporated into the model calibration. However, it is rare to have discharge measurement data when catchments are ungauged. In this study, we incorporate discharge estimates derived from regionalization into a water-level based calibration using an inverse rating curve (IRC) method. Specifically, discharge indices such as the 95th percentile of daily discharge (Q95) and the mean daily discharge (Qmean) are estimated from catchment and climate characteristics based on regionalized relationships. These estimated discharge indices are then used to constrain the IRC calibration. We evaluate both constrained and unconstrained IRC methods over 130 study catchments where observed discharge time series are available but used only for validation in this study. Modelled discharge time series from the constrained IRC methods have median Nash Sutcliffe Efficiency (NSE) values of 0.61–0.63, compared with 0.54 from the unconstrained IRC method. Improvement is seen in around 75% of the catchments, with more than 0.2 increase in NSE values in 25% of the catchments. Improvement in the modelled discharges from using regionalized discharge indices is comparable to using observed discharge indices. Using regionalized Qmean yields greater improvement than using regionalized Q95. The constrained IRC methods tend to perform better in wetter catchments than in drier ones. This method can significantly extend rainfall-runoff model calibration in a wider range of catchments, including real ungauged catchments. [ABSTRACT FROM AUTHOR]

Floodplains are a key ecological feature of arid and semi-arid regions and flooding is the main source of vegetation productivity. Characterizing the spatiotemporal flood dynamics, such as surface water and changes in subsequent vegetation growth vigour and biomass, is essential in better understanding of ecology and hydrology of these regions. Recent remotely sensed flooding studies of arid zone floodplains have concentrated on improving classification of surface water, particularly in mixed water-vegetation areas; less is known about the effect of surface water flooding pattern on emergent vegetation dynamics. We use an integrated framework for mapping both flood extents and the persistence of floodplain response changes of water, vegetation and moisture in Cooper Creek, Australia. We analysed pixel-based time series of multiple indices generated by daily MODIS data for 14 highly variable flow flood events between 2000 and 2012. Results indicate that for the extremely flat Cooper Creek floodplain, mapping inundation area by changes in vegetation growth vigour and biomass was significantly larger than surface water mapping area (16.2% for the 2004 flood event) and the difference in inundation mapping mainly occurs around the inundated edges. In addition, by studying surface water and subsequent vegetation response together, it is possible to generate new information, such as the lag time between flooding and peak vegetation growth vigour and biomass, and persistence time of surface water and green vegetation, which provide important insights to arid zone floodplain behaviour. The large extent and high frequency of MODIS images provide advantages in characterizing inundation dynamics for large-scale floodplains where instantaneous (daily) inundation extent is considerably smaller than total cumulative inundation extent, compared with sensors with higher spatial resolution but lower temporal resolution; however, the coarse resolution of MODIS (500 m) limits its performance for small flood events. Globally, this approach is suitable for other large, low-gradient floodplains in arid zones that show similar, long duration vegetation responses as observed in Cooper Creek. [ABSTRACT FROM AUTHOR]

Active and passive microwave satellite remote sensing are enabling sub-daily global observations of surface soil moisture (SM) for hydrological, meteorological and climatological studies. Because the retrieved SM data can be quite noisy, post-retrieval processing such as de-noising can play an important role to aid interpretation of the observed dynamics or enhance their utility for data assimilation. To date, the merits of such techniques have not yet been fully evaluated. Here we consider the applications of Fourier-based de-noising filters of Su et al. (2013a) for improving SM retrieved by AMSR-E (Advanced Microwave Scanning Radiometer for Earth Observing System) and ASCAT (Advanced Scatterometer of MetOp-A) sensors. The filters are calibrated in the frequency domain based on a water-balance model, without the need for ancillary data. The evaluation of the de-noising methods was conducted globally against in situ data distributed via the International Soil Moisture Network (ISMN) at 277 AMSR-E and 385 ASCAT pixels. Systematic improvements were found for all considered metrics, namely root-mean-square deviation, linear correlation and signal-to-noise ratio, for both SM products, with improvements more striking for AMSR-E. However, the originally proposed implementation of the filters can induce undesirable over-smoothing and distortion of SM timeseries. To overcome this, based on a simple heuristic argument, we propose the use of ancillary precipitation data in the filtering process, although at some expense of overall agreements with the in situ data. [ABSTRACT FROM AUTHOR]