Downscaling rainfields in space and time, using the String of Beads model in time series mode

The String of Beads model is a space-time model of rain fields measured by weather radar. It is here driven by two auto-regressive time series models. one at the image scale, the other at the pixel scale. to model the tempora l correlation structure of the wet-period process. The marginal distributio n of the pixel scale intensities on a given radar-rainfall image is describ ed by a log-normal distribution. The spatial dependence structure of each i mage is defined by a power spectrum approximated by a power law function wi th a negative exponent, It is demonstrated that this stochastic modelling a pproach is valid because the images sampled are effectively stationary abov e a scale of 30 km, which is less than a quarter of the image width. By adv ecting a simulated sequence of images along the same cumulative advection v ector as the observed event and matching the image-scale statistics of each simulated image with those of the corresponding observed image, a simulate d sequence of plausible images is generated which mimics (has the same spac e-time statistics as) the observed event but differs from it in detail. Agg regating the pixel scale intensities in each sequence over a number of time and space intervals and then comparing their spatial and temporal statisti cs, demonstrates that the model captures the intermediate scale behaviour w ell, showing satisfactorily its ability to downscale rainfall in space and time. The model thus has potential as an operational space-time model of ra infields.