Weather radar has been widely used in hydrologic forecasting and decision making; nevertheless, there is increasing attention on its unceltainties that propagates through hydrologic models. These unceltainties are not only caused by radar itself in measuring and estimating rainfall (such as attenuation, extrapolation of the rainfall measured aloft to the ground, sampling methods and pmtial beam blocking, etc.), but also come from the complicated synoptic regimes (such as air motion, vertical variability of temperature, and conversion to and from different hydrometeors). This thesis aims to improve the quality of radar rainfall and describe its uncertainty. The Brue catchment (135 sq. km) in Southwest England covering 28 radar pixels and 49 rain gauges and a hilly area to the east and south of Manchester with around 5000 sq . km and 50 rain gauges are chosen as the experimental domains for this thesis. The studies are composed of three main pmts: Firstly, I propose a fully formulated uncertainty model that can statistically quantify the characteristics of the radar rainfall errors and their spatial and temporal structure, which is a novel method of its kind in the radar data unceltainty field. The uncertainty model is established based on the distribution of gauge rainfall conditioned on radar rainfall (GRJRR). Its spatial and temporal dependences are simulated based on the copula function. With this proposed uncertainty model, a Multivariate Distributed Ensemble Generator (MDEG) driven by the copula and autoregressive filter is designed. The products from MDEG include a time series of ensemble rainfall fields with each of them representing a probable true rainfall. As wind is a typical weather factor that influences radar measurement, this thesis introduces the wind field into the unceltainty model and designs the radar rainfall uncertainty model under different wind conditions. In addition, I also propose an Empirically-based Ensemble Rainfall Forecasts (ERFEM) model to measure and quantify the combined effect of all the error sources in the radarrainfall forecasts .. The essence of the unceltainty model is formulated into an empirical relation between the radar rainfall forecasts and the corresponding 'ground truth' represented by the rainfall field from rain gauges. In modelling the radar rainfall unceltainty, I find that the wind has a huge impact on radar-gauge comparison. Due to the wind effects, the raindrops observed by the radar do not always fall veltically to the ground, and the raindrops arriving at the ground cannot all be caught by the rain gauge. This thesis proposes a practical approach to simulate the movement of raindrops in the air and adjust the aforementioned wind-induced errors on radar bias correction procedure. This scheme is based on the numerical simulation of raindrop movements in the three-dimensional ' wind field. The Weather Research and Forecasting (WRF) model is used to downs.cale the reanalysis data ERA-40 to obtain the wind field with high spatial and temporal resolutions. A normalized gamma model is adopted to estimate the raindrop size distribution (DSD). This work is the first study to tackle both wind effects on radar and rain gauges, which could be considered as one of the essential components in processing radar observation data, which should be undertaken after the aforementioned physical processes and before bias correction. Finally, this thesis analyzes how the radar rainfall uncertainty propagates through a hydrological model (the Xinanjiang model) and investigates which features of the uncertainty model have significant impacts on flow simulation. The generated ensemble rainfall values by MDEG are input into the Xinanjiang model to produce uncertainty bands of the ensemble flows. Five important indicators are used to describe the characteristics of uncertainty bands. It is concluded that the Gaussian marginal distribution and spatio-temporal dependence using Gaussian copula is considered to be the preferred configuration of the MDEG model for hydrological model unceltainty analysis in the Brue catchment. Keywords: Multivariate Distributed Ensemble Generator (MDEG); Copula; flow simulation; Radar-Rainfall Estimates; Hydrological Model Uncertainty; wind-induced error; drop size distribution; WRF; wind-drift.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:681555 |
| Date | January 2014 |
| Creators | Dai, Qiang |
| Publisher | University of Bristol |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0243 seconds