Toward an Ensemble Radar Precipitation Estimation Based on a Dynamic Description of the Measurement Errors

Radar based Quantitative Precipitation Estimation (QPE) is increasingly being used to initialise hydrological models. Radars can provide higher temporal and spatial resolution observations of rainfall than rain-gauges and are critical instruments for predicting flash flood events (Vincendon et al, 2010). However, radar based QPE is affected by retrieval errors that are an important source of uncertainty for hydrological predictions.

Radar ensembles have been proposed by numerous authors to represent the precipitation estimate as well as the uncertainty. Each member of a radar ensemble is an estimate of the unknown and true precipitation field and the spread of the ensemble a measure of the uncertainty.

Radar ensembles can be generated either by modelling individual sources of error, (Lee et al., (2007), Lee and Zawadzki (2005a, 2005b, 2006), Jordan et al (2003), Berenguer and Zawadzki (2008)) or by describing statistically the residual errors in radar based QPE (Germann et al (2009), Ciach et al (2007), Llort et al (2008)). As explained by Seed et al, (2013), the difficulty with the first approach is that error structure is complex and inter-dependent. The challenge with the later approach is the requirement for a reference field: this is usually a dense network of point observations from rain gauges.

In this study, we chose to statistically study the errors of the radar based QPE, as in Germann et al.(2009). The original method is based on singular value decomposition of the error covariance matrix, stochastic simulation using the LU decomposition algorithm, and autoregressive filtering. But, in order to take into account different sources of error (Z-R relationships, VPR correction, distance from the radar), we produced separate statistics and error covariance matrices for different classes of precipitation intensity, freezing level height and quality index, since the quality index decreases with both the distance from the radar and the height of the beam above the ground, as explained in Tabary et al. (2007). For each parameter, three classes have been considered and therefore 27 error covariance matrices are used in the ensemble generator to produce perturbations dynamically linked to the radar measurements.

The results presented at the conference will show to which extend the new method can be used to both improve QPE estimates and provide and provide a more realistic estimation of the uncertainty.

Berenguer, M., & Zawadzki, I. (2008). A study of the error covariance matrix of radar rainfall estimates in stratiform rain. Weather and Forecasting, 23(6), 1085-1101.

Ciach, G. J., Krajewski, W. F., & Villarini, G. (2007). Product-error-driven uncertainty model for probabilistic quantitative precipitation estimation with NEXRAD data. Journal of Hydrometeorology, 8(6), 1325-1347.

Germann, U., Berenguer, M., Sempere‐Torres, D., & Zappa, M. (2009). REAL—Ensemble radar precipitation estimation for hydrology in a mountainous region. Quarterly Journal of the Royal Meteorological Society, 135(639), 445-456.

Jordan, P. W., Seed, A. W., & Weinmann, P. E. (2003). A stochastic model of radar measurement errors in rainfall accumulations at catchment scale. Journal of Hydrometeorology, 4(5), 841-855.

Lee, G. W., Seed, A. W., & Zawadzki, I. (2007). Modeling the variability of drop size distributions in space and time. Journal of applied meteorology and climatology, 46(6), 742-756.

Lee, G. W., & Zawadzki, I. (2005a). Variability of drop size distributions: Time-scale dependence of the variability and its effects on rain estimation. Journal of applied meteorology, 44(2), 241-255.

Lee, G. W., & Zawadzki, I. (2005b). Variability of drop size distributions: Noise and noise filtering in disdrometric data. Journal of applied meteorology, 44(5), 634-652.

Lee, G., & Zawadzki, I. (2006). Radar calibration by gage, disdrometer, and polarimetry: Theoretical limit caused by the variability of drop size distribution and application to fast scanning operational radar data. Journal of Hydrology, 328(1), 83-97.

Llort, X., Velasco-Forero, C., Roca-Sancho, J., & Sempere-Torres, D. (2008, June). Characterization of uncertainty in radar-based precipitation estimates and ensemble generation. In Fifth European Conference on Radar in Meteorology and Hydrology (Vol. 12).

Seed, A. W., Pierce, C. E., & Norman, K. (2013). Formulation and evaluation of a scale decomposition‐based stochastic precipitation nowcast scheme. Water Resources Research, 49(10), 6624-6641.

Tabary, P. (2007). The new French operational radar rainfall product. Part I: Methodology. Weather and forecasting, 22(3), 393-408.

Tabary, P., Desplats, J., Do Khac, K., Eideliman, F., Gueguen, C., & Heinrich, J. C. (2007). The new French operational radar rainfall product. Part II: Validation. Weather and forecasting, 22(3), 409-427.

Vincendon, B., Ducrocq, V., Saulnier, G. M., Bouilloud, L., Chancibault, K., Habets, F., & Noilhan, J. (2010). Benefit of coupling the ISBA land surface model with a TOPMODEL hydrological model version dedicated to Mediterranean flash-floods. Journal of Hydrology, 394(1), 256-266.