The Hydrologist's traditional tool for measuring rainfall is the rain gauge. Rain gauges are relatively cheap, easy to maintain and provide a direct and suitably accurate estimate of rainfall at a point. What rain gauges fail to capture well is the spatial variability of rainfall with time, an important aspect for the credible modelling of a catchment's response to rainfall. This spatial variability is particularly evident at short timescales of up to several days. As the period of accumulation increases, the expected spatial variability is reduced and rain gauges provide improved spatial rainfall estimates. Because of the fractal variability of rainfall in space, simple interpolation between rain gauges does not provide an accurate estimate of the true spatial rainfall field, at short time scales.Weather radar provides (with a single instrument) a highly detailed representation of the spatial structure and temporal evolution of rainfall over a large area. Estimated rainfall rates are derived indirectly from measurements of reflectivity and are therefore subject to a combination of systematic and random errors.This article describes a recently proposed merging technique and presents an application to simulated rainfall fields. The technique employed is Conditional Merging (Ehret, 2002), which makes use of Kriging to extract the optimal information content from the observed data. A mean field based on the Kriged rain gauge data is adopted, while the spatial detail from the radar is retained, reducing bias, but keeping the spatial variability observed by the radar. The variance of the estimate is reduced in the vicinity of the gauges where they are able to provide good information on the true rainfall field.

Abstract. The paper compares two independent approaches to estimate soil moisture at the regional scale over a 4625 km 2 catchment (Liebenbergsvlei, South Africa). The first estimate is derived from a physically-based hydrological model (TOPKAPI). The second estimate is derived from the scatterometer on board the European Remote Sensing satellite (ERS). Results show a good correspondence between the modelled and remotely sensed soil moisture, particularly with respect to the soil moisture dynamic, illustrated over two selected seasons of 8 months, yielding regression R 2 coefficients lying between 0.68 and 0.92. Such a close similarity between these two different, independent approaches is very promising for (i) remote sensing in general (ii) the use of hydrological models to back-calculate and disaggregate the satellite soil moisture estimate and (iii) for hydrological models to assimilate the remotely sensed soil moisture.

In this paper we compare two independent soil moisture estimates over South Africa. The first estimate is a Soil Saturation Index (SSI) provided by automated real-time computations of the TOPKAPI hydrological model, adapted to run as a collection of independent 1 km cells with centres on a grid with a spatial resolution of 0.125 • , at 3 h intervals. The second set of estimates is the remotely sensed ASCAT Surface Soil Moisture product, temporally filtered to yield a Soil Wetness Index (SWI). For the TOPKAPI cells, the rainfall forcing used is the TRMM 3B42RT product, while the evapotranspiration forcing is based on a modification of the FAO56 reference crop evapotranspiration (ET 0). ET 0 is computed using forecast fields of meteorological variables from the Unified Model (UM) runs done by the South African Weather Service (SAWS); the UM forecast fields were used, because reanalysis is not done by SAWS. To validate these ET 0 estimates we compare them with those computed using observed meteorological data at a network of weather stations ; they were found to be unbiased with acceptable scatter. Using the rainfall and evapotranspiration forcing data, the percentage saturation of the TOPKAPI soil store is computed as a Soil Saturation Index (SSI), for each of 6984 uncon-nected uncalibrated TOPKAPI cells at 3 h time-steps. These SSI estimates are then compared with the SWI estimates obtained from ASCAT. The comparisons indicate a good correspondence in the dynamic behaviour of SWI and SSI for a significant proportion of South Africa.

Abstract. In this paper we compare two independent soil moisture estimates over South Africa. The first estimate is a Soil Saturation Index (SSI) provided by automated real-time computations of the TOPKAPI hydrological model, adapted to run as a collection of independent 1 km cells with centres on a grid with a spatial resolution of 0.125 • , at 3 h intervals. The second set of estimates is the remotely sensed ASCAT Surface Soil Moisture product, temporally filtered to yield a Soil Wetness Index (SWI). For the TOPKAPI cells, the rainfall forcing used is the TRMM 3B42RT product, while the evapotranspiration forcing is based on a modification of the FAO56 reference crop evapotranspiration (ET 0 ). ET 0 is computed using forecast fields of meteorological variables from the Unified Model (UM) runs done by the South African Weather Service (SAWS); the UM forecast fields were used, because reanalysis is not done by SAWS. To validate these ET 0 estimates we compare them with those computed using observed meteorological data at a network of weather stations; they were found to be unbiased with acceptable scatter. Using the rainfall and evapotranspiration forcing data, the percentage saturation of the TOPKAPI soil store is computed as a Soil Saturation Index (SSI), for each of 6984 unconnected uncalibrated TOPKAPI cells at 3 h time-steps. These SSI estimates are then compared with the SWI estimates obtained from ASCAT. The comparisons indicate a good correspondence in the dynamic behaviour of SWI and SSI for a significant proportion of South Africa.

Abstract.A data-driven method for extracting temporally persistent information, at different spatial scales, from rainfall data (as measured by radar/satellite) is described, which extends the Empirical Mode Decomposition (EMD) algorithm into two dimensions. The EMD technique is used here to decompose spatial rainfall data into a sequence of high through to low frequency components. This process is equivalent to the application of successive low-pass spatial filters, but based on the observed properties of the data rather than the predetermined basis functions used in traditional Fourier or Wavelet decompositions. It has been suggested in the literature that the lower frequency components (those with large spatial extent) of spatial rainfall data exhibit greater temporal persistence than the higher frequency ones. This idea is explored here in the context of Empirical Mode Decomposition. The paper focuses on the implementation and development of the two-dimensional extension to the EMD algorithm and it's application to radar rainfall data, as well as examining temporal persistence in the data at different spatial scales.