Dynamical and Microphysical Retrieval from Simulated Doppler Radar Observations Using the 4DVAR Assimilation Technique

Based on a cloud model and the four-dimensional variational (4DVAR) dataassimilation method developed by Sun and Crook (1997), simulated experiments of dynamical andmicrophysical retrieval from Doppler radar data were performed. The 4DVAR data assimilationtechnique was applied to a cloud scale model with a warm rain parameterization scheme. The 3D wind,thermodynamical, and microphysieal fields were determined by minimizing a cost function, defined bythe difference between both radar observed radial velocities and reflectivities and their modelpredictions. The adjoint of the numerical model was used to provide the gradient of the costfunction with respect to the control variables. Experiments have demonstrated that the 4DVARassimilation method is able to retrieve the detailed structure of wind, thermodynamics, andmicrophysics by using either dual-Doppler or single-Doppler information. The quality of retrievaldepends strongly on the magnitude of constraint with respect to the variables. Retrieving thetemperature field, cloud water and water vapor is more difficult than the recovery of the wind fieldand rainwater. Accurate thermodynamic retrieval requires a longer assimilation period. Theinclusion of a background term, even mean fields from a single sounding, helped reduce the retrievalerrors. Less accurate velocity fields were obtained when single-Doppler data were used. It wasfound that the retrieved velocity is sensitive to the location of the retrieval domain relative tothe radars while the other fields have very little changes. Two radar volumetric scans are generallyadequate for providing the evolution, although the use of additional volumes improves theretrieval. As the amount of the observations decreases, the performance of the retrieval isdegraded. However, the missing observations can be compensated by adding a background term to thecost function. The technique is robust to random errors in radial velocity and calibration errors inreflectivity. The boundary conditions from the dual-Doppler synthesized winds are sufficient forthe retrieval. When the retrieval is mainly controlled by the observations in the regions away fromthe boundaries, the simple boundary conditions from velocity azimuth display (VAD) analysis are alsoavailable. The microphysical retrieval is sensitive to model errors.

Mesoscale Structure of Rainstorm Retrieved from Dual-Doppler Radar Observations Using the 4DVAR Assimilation Technique

The four-dimensional variational （4DVAR） data assimilation method was applied to dual-Doppler radar data about two Meiyu rainstorms observed during CHeRES （China Heavy Rain Experiment and Study）. The purpose of this study is to examine the performance of the 4DVAR technique in retrieving rainstorm mesoscale structure and to reveal the feature of rainstorm mesoscale structure. Results demonstrated that the 4DVAR assimilation method was able to retrieve the detailed structure of wind, thermodynamics, and microphysics fields from dual-Doppler radar observations. The retrieved wind fields agreed with the dual- Doppler synthesized winds and were accurate. The distributions of the retrieved perturbation pressure, perturbation temperature, and microphysics fields were also reasonable through the examination of their physical consistency. Both of the two heavy rainfalls were caused by merging cloud processes. The wind shear and convergence lines at middle and lower levels were their primary dynamical characteristics. The convective system was often related to low-level convergence and upper-level divergence coupled with up- drafts. During its mature stage, the convective system was characterized by low pressure at lower level and high pressure at upper level, associated with warmer at middle level and colder at lower and upper levels than the environment. However, a region of cooling and high pressure occurred in the lower and middle levels compared to warming and low pressure in the upper level during its dissipating ＇.stage. The water vapor, cloud water, and rainwater corresponded to the convergence, the updraft and the intensive reflectivity, respectively.

Toward development of the 4Dvar data assimilation system in the Bering Sea:reconstruction of the mean dynamic ocean topography

The Bering Sea circulation is derived as a variational inverse of hydrographic profiles（ temperature and salinity） , atmospheric climatologies and historical observation of ocean curents. The important result of this study is estimate of the mean climatological sea surface height （SSH） that can be used as a reference for satellite altimetry sea level anomaly data in the Bering Sea region. Numerical experiments reveal that, when combined with satellite altimetry, the obtained reference SSH effectively constrains a realistic reconstruction of the Amukta Pass circulation.

Three-dimensional structure of an observed cyclonic mesoscale eddy in the Northwest Pacific and its assimilation experiment

Mesoscale eddies play an important role in modulating the ocean circulation.Many previous studies on the threedimensional structure of mesoscale eddies were mainly based on composite analysis,and there are few targeted observations for individual eddies.A cyclonic eddy surveyed during an oceanographic cruise in the Northwest Pacific Ocean is investigated in this study.The three-dimensional structure of this cyclonic eddy is revealed by observations and simulated by the four-dimensional variational data assimilation(4 DVAR)system combined with the Regional Ocean Modeling System.The observation and assimilation results together present the characteristics of the cyclonic eddy.The cold eddy has an obvious dual-core structure of temperature anomaly.One core is at 50–150 m and another is at 300–550 m,which both have the average temperature anomaly of approximately-3.5℃.The salinity anomaly core is between 250 m and 500 m,which is approximately-0.3.The horizontal velocity structure is axis-asymmetric and it is enhanced on the eastern side of the cold eddy.In the assimilation experiment,sea level anomaly,sea surface temperature,and in situ measurements are assimilated into the system,and the results of assimilation are close to the observations.Based on the high-resolution assimilation output results,the study also diagnoses the vertical velocity in the mesoscale eddy,which reaches the maximum of approximately 10 m/d.The larger vertical velocity is found to be distributed in the range of 0.5 to 1 time of the normalized radius of the eddy.The validation of the simulation result shows that the 4 DVAR method is effective to reconstruct the three-dimensional structure of mesoscale eddy and the research is an application to study the mesoscale eddy in the Northwest Pacific by combining observation and assimilation methods.

Rainfall Assimilation Using a New Four-Dimensional Variational Method:A Single-Point Observation Experiment

Accurate forecast of rainstorms associated with the mei-yu front has been an important issue for the Chinese economy and society. In July 1998 a heavy rainstorm hit the Yangzi River valley and received widespread attention from the public because it caused catastrophic damage in China. Several numerical studies have shown that many forecast models, including Pennsylvania State University National Center for Atmospheric Research’s fifth-generation mesoscale model （MM5）, failed to simulate the heavy precipitation over the Yangzi River valley. This study demonstrates that with the optimal initial conditions from the dimension-reduced projection four-dimensional variational data assimilation （DRP-4DVar） system, MM5 can successfully reproduce these observed rainfall amounts and can capture many important mesoscale features, including the southwestward shear line and the low-level jet stream. The study also indicates that the failure of previous forecasts can be mainly attributed to the lack of mesoscale details in the initial conditions of the models.

The REM Adjoint System and Its 4DVar Data Assimilation Experiments

The Regional Eta-coordinate Model（REM） has performed well in forecasting heavy rainfalls in China in recent years.A four-dimensional variational assimilation system（4DVar） is developed to improve the forecast skill of the REM.The tangent linear model and adjoint model codes are written according to the＂code to code＂rule,and the establishment of the REM adjoint modeling system is introduced in detail in this paper.The tangent linear and adjoint models of the REM are validated against the observational data,and so is the gradient of the given cost function.It is shown that for the tangent linear model and cost function,when the magnitude of perturbations is reduced,the verification results approach 1.0;when the rounding error of computer is increased,the verification results depart off 1.0.In the validation of the adjoint model,the values on the left- and right-hand sides of the algebraic formula are equal with 13-digit accuracy.These results indicate that the tangent linear model and the adjoint model system of the REM are successfully coded,and the gradient of the cost function is correctly calculated.By using the REM adjoint modeling system,two 4DVar experiments and extended forecasts are performed using observational data for two real cases in June 1998 and August 2000.The results show that forecasts of temperature,wind speed, and specify humidity using the 4DVar-assimilated initial data are all improved at the end of the forecast period.However,the performance of the 4DVar in forcasting rainfall is different in these two cases.The prediction of location and amount of the accumulated rainfall is well improved in the first case,while in the second case the prediction has no significant improvement.The problem may result from the fact that the observational data used in the 4DVar for the second case are inadequate.This case will be studied further in future work.