[1] Rain gauge networks provide rainfall measurements with a high degree of accuracy at specific locations but, in most cases, the instruments are too sparsely distributed to accurately capture the high spatial and temporal variability of precipitation systems. Radar and satellite remote sensing of rainfall has become a viable approach to address this problem effectively. However, among other sources of uncertainties, the remote-sensing based rainfall products are unavoidably affected by sampling errors that need to be evaluated and characterized. Using a large data set (more than six years) of rainfall measurements from a dense network of 50 rain gauges deployed over an area of about 135 km 2 in the Brue catchment (south-western England), this study sheds some light on the temporal and spatial sampling uncertainties: the former are defined as the errors resulting from temporal gaps in rainfall observations, while the latter as the uncertainties due to the approximation of an areal estimate using point measurements. It is shown that the temporal sampling uncertainties increase with the sampling interval according to a scaling law and decrease with increasing averaging area with no strong dependence on local orography. On the other hand, the spatial sampling uncertainties tend to decrease for increasing accumulation time, with no strong dependence on location of the gauge within the pixel or on the gauge elevation. For the evaluation of high resolution satellite rainfall products, a simple rule is proposed for the number of rain gauges required to estimate areal rainfall with a prescribed accuracy. Additionally, a description is given of the characteristics of the rainfall process in the area in terms of spatial correlation.

A novel analogue-based heuristic tool for nowcasting orographic precipitation is presented. The system takes advantage of the orographic forcing, which determines a strong relation between mesoscale flows, air mass stability and rainfall patterns. These quantities are used as predictors of precipitation. In particular, past situations with the predictors most similar to those observed at the current time are identified by searching a large historical dataset. Deterministic and probabilistic forecasts are then generated every five minutes as new observations are available, based on the rainfall observed by radar after the analogous situations. The analogue method provides a natural way to incorporate evolution of precipitation into the nowcasting system and to express forecast uncertainty by means of ensembles.A total of 127 days of long-lasting orographic precipitation constitutes the historical dataset in which the analogous situations are searched. The system is developed for the Lago Maggiore region in the southern part of the European Alps. Given the availability of radar data and the presence of a strong orographic forcing, it can be extended to other mountainous regions. An evaluation of the skill of the system shows that the heuristic tool performs better than Eulerian persistence for predictions with lead time larger than one hour, and better than the numerical model COSMO2 for forecasts with lead time up to four hours.

No abstract

[1] The goal of this study is to diagnose the manner in which radar-rainfall input affects peak flow simulation uncertainties across scales. We used the distributed physically based hydrological model CUENCAS with parameters that are estimated from available data and without fitting the model output to discharge observations. We evaluated the model's performance using (1) observed streamflow at the outlet of nested basins ranging in scale from 20 to 16,000 km 2 and (2) streamflow simulated by a well-established and extensively calibrated hydrological model used by the US National Weather Service (SAC-SMA). To mimic radar-rainfall uncertainty, we applied a recently proposed statistical model of radar-rainfall error to produce rainfall ensembles based on different expected error scenarios. We used the generated ensembles as input for the hydrological model and summarized the effects on flow sensitivities using a relative measure of the ensemble peak flow dispersion for every link in the river network. Results show that peak flow simulation uncertainty is strongly dependent on the catchment scale. Uncertainty decreases with increasing catchment drainage area due to the aggregation effect of the river network that filters out small-scale uncertainties. The rate at which uncertainty changes depends on the error structure of the input rainfall fields. We found that random errors that are uncorrelated in space produce high peak flow variability for small scale basins, but uncertainties decrease rapidly as scale increases. In contrast, spatially correlated errors produce less scatter in peak flows for small scales, but uncertainty decreases slowly with increasing catchment size. This study demonstrates the large impact of scale on uncertainty in hydrological simulations and demonstrates the need for a more robust characterization of the uncertainty structure in radar-rainfall. Our results are diagnostic and illustrate the benefits of using the calibrationfree, multiscale framework to investigate uncertainty propagation with hydrological models.Citation: Cunha, L. K., P. V. Mandapaka, W. F. Krajewski, R. Mantilla, and A. A. Bradley (2012), Impact of radar-rainfall error structure on estimated flood magnitude across scales: An investigation based on a parsimonious distributed hydrological model, Water Resour. Res., 48, W10515,

No abstract