Land surface temperature(LST)is the key variable in land-atmosphere interaction,having an important impact on weather and climate forecasting.However,achieving consistent analysis of LST and the atmosphere in assimila...Land surface temperature(LST)is the key variable in land-atmosphere interaction,having an important impact on weather and climate forecasting.However,achieving consistent analysis of LST and the atmosphere in assimilation is quite challenging.This is because there is limited knowledge about the cross-component background error covariance(BEC)between LST and atmospheric state variables.This study aims to clarify whether there is a relationship between the error of LST and atmospheric variables,and whether this relationship varies spatially and temporally.To this end,the BEC coupled with atmospheric variables and LST was constructed(LST-BEC),and its characteristics were analyzed based on the 2023 mei-yu season.The general characteristics of LST-BEC show that the LST is mainly correlated with the atmospheric temperature and the correlation decreases gradually with a rise in atmospheric height,and the error standard deviation of the LST is noticeably larger than that of the low-level atmospheric temperature.The spatiotemporal characteristics of LST-BEC on the heavy-rain day and light-rain day show that the error correlation and error standard deviation of LST and low-level atmospheric temperature and humidity are closely related to the weather background,and also have obvious diurnal variations.These results provide valuable information for strongly coupled land-atmosphere assimilation.展开更多
Background error covariance plays an important role in any variational data assimilation system, because it determines how information from observations is spread in model space and between different model variables. ...Background error covariance plays an important role in any variational data assimilation system, because it determines how information from observations is spread in model space and between different model variables. In this paper, the use of orthogonal wavelets in representation of background error covariance over a limited area is studied. Based on the WRF model and its 3D-VAR system, an algorithm using orthogonal wavelets to model background error covariance is developed. Because each wavelet function contains information on both position and scale, using a diagonal correlation matrix in wavelet space gives the possibility to represent some anisotropic and inhomogeneous characteristics of background error covariance. The experiments show that local correlation functions are better modeled than spectral methods. The formulation of wavelet background error covariance is tested with the typhoon Kaemi (2006). The results of experiments indicate that the subsequent forecasts of typhoon Kaemi’s track and intensity are significantly improved by the new method.展开更多
The quality of background error statistics is one of the key components for successful assimilation of observations in a numerical model.The background error covariance(BEC) of ocean waves is generally estimated under...The quality of background error statistics is one of the key components for successful assimilation of observations in a numerical model.The background error covariance(BEC) of ocean waves is generally estimated under an assumption that it is stationary over a period of time and uniform over a domain.However,error statistics are in fact functions of the physical processes governing the meteorological situation and vary with the wave condition.In this paper,we simulated the BEC of the significant wave height(SWH) employing Monte Carlo methods.An interesting result is that the BEC varies consistently with the mean wave direction(MWD).In the model domain,the BEC of the SWH decreases significantly when the MWD changes abruptly.A new BEC model of the SWH based on the correlation between the BEC and MWD was then developed.A case study of regional data assimilation was performed,where the SWH observations of buoy 22001 were used to assess the SWH hindcast.The results show that the new BEC model benefits wave prediction and allows reasonable approximations of anisotropy and inhomogeneous errors.展开更多
Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor c...Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor control variables are necessary.Common data assimilation systems theoretically require that the probability density functions(PDFs)of analysis,background,and observation errors should satisfy the Gaussian unbiased assumptions.In this study,a Gaussian transform method is proposed to transform hydrometeors to more Gaussian variables,which is modified from the Softmax function and renamed as Quasi-Softmax transform.The Quasi-Softmax transform method then is compared to the original hydrometeor mixing ratios and their logarithmic transform and Softmax transform.The spatial distribution,the non-Gaussian nature of the background errors,and the characteristics of the background errors of hydrometeors in each method are studied.Compared to the logarithmic and Softmax transform,the Quasi-Softmax method keeps the vertical distribution of the original hydrometeor mixing ratios to the greatest extent.The results of the D′Agostino test show that the hydrometeors transformed by the Quasi-Softmax method are more Gaussian when compared to the other methods.The Gaussian transform has been added to the control variable transform to estimate the background error covariances.Results show that the characteristics of the hydrometeor background errors are reasonable for the Quasi-Softmax method.The transformed hydrometeors using the Quasi-Softmax transform meet the Gaussian unbiased assumptions of the data assimilation system,and are promising control variables for data assimilation systems.展开更多
Satellite data obtained over synoptic data-sparse regions such as an ocean contribute toward improving the quality of the initial state of limited-area models. Background error covariances are crucial to the proper di...Satellite data obtained over synoptic data-sparse regions such as an ocean contribute toward improving the quality of the initial state of limited-area models. Background error covariances are crucial to the proper distribution of satellite-observed information in variational data assimilation. In the NMC （National Meteorological Center） method, background error covariances are underestimated over data-sparse regions such as an ocean because of small differences between different forecast times. Thus, it is necessary to reconstruct and tune the background error covariances so as to maximize the usefulness of the satellite data for the initial state of limited-area models, especially over an ocean where there is a lack of conventional data. In this study, we attempted to estimate background error covariances so as to provide adequate error statistics for data-sparse regions by using ensemble forecasts of optimal perturbations using bred vectors. The background error covariances estimated by the ensemble method reduced the overestimation of error amplitude obtained by the NMC method. By employing an appropriate horizontal length scale to exclude spurious correlations, the ensemble method produced better results than the NMC method in the assimilation of retrieved satellite data. Because the ensemble method distributes observed information over a limited local area, it would be more useful in the analysis of high-resolution satellite data. Accordingly, the performance of forecast models can be improved over the area where the satellite data are assimilated.展开更多
Background error covariance(BEC)plays an essential role in variational data assimilation.Most variational data assimilation systems still use static BEC.Actually,the characteristics of BEC vary with season,day,and eve...Background error covariance(BEC)plays an essential role in variational data assimilation.Most variational data assimilation systems still use static BEC.Actually,the characteristics of BEC vary with season,day,and even hour of the background.National Meteorological Center-based diurnally varying BECs had been proposed,but the diurnal variation characteristics were gained by climatic samples.Ensemble methods can obtain the background error characteristics that suit the samples in the current moment.Therefore,to gain more reasonable diurnally varying BECs,in this study,ensemble-based diurnally varying BECs are generated and the diurnal variation characteristics are discussed.Their impacts are then evaluated by cycling data assimilation and forecasting experiments for a week based on the operational China Meteorological Administration-Beijing system.Clear diurnal variation in the standard deviation of ensemble forecasts and ensemble-based BECs can be identified,consistent with the diurnal variation characteristics of the atmosphere.The results of one-week cycling data assimilation and forecasting show that the application of diurnally varying BECs reduces the RMSEs in the analysis and 6-h forecast.Detailed analysis of a convective rainfall case shows that the distribution of the accumulated precipitation forecast using the diurnally varying BECs is closer to the observation than using the static BEC.Besides,the cycle-averaged precipitation scores in all magnitudes are improved,especially for the heavy precipitation,indicating the potential of using diurnally varying BEC in operational applications.展开更多
In the previous study, the influences of introducing larger- and smaller-scale errors on the background error covariances estimated at the given scales were investigated, respectively. This study used the eovariances ...In the previous study, the influences of introducing larger- and smaller-scale errors on the background error covariances estimated at the given scales were investigated, respectively. This study used the eovariances obtained in the previous study in the data assimilation and model forecast system based on three-dimensional variational method and the Weather Research and Forecasting model. In this study, analyses and forecasts from this system with different covariances for a period of one month were compared, and the causes for differing results were presented. The varia- tions of analysis increments with different-scale errors are consistent with those of variances and correlations of back- ground errors that were reported in the previous paper. In particular, the introduction of smaller-scale errors leads to greater amplitudes in analysis increments for medium-scale wind at the heights of both high- and low-level jets. Tem- perature and humidity analysis increments are greater at the corresponding scales at the middle- and upper-levels. These analysis increments could improve the intensity of the jet-convection system that includes jets at different levels and the coupling between them that is associated with latent heat release. These changes in analyses will contribute to more ac- curate wind and temperature forecasts in the corresponding areas. When smaller-scale errors are included, humidity analysis increments are significantly enhanced at large scales and lower levels, to moisten southern analyses. Thus, dry bias can be corrected, which will improve humidity forecasts. Moreover, the inclusion of larger- （smaller-） scale errors will be beneficial for the accuracy of forecasts of heavy （light） precipitation at large （small） scales because of the ampli- fication （diminution） of the intensity and area in precipitation forecasts.展开更多
Based on the optimal interpolation objective analysis of the Argo data, improvements are made to the em- pirical formula of a background error covariance matrix widely used in data assimilation and objective anal- ysi...Based on the optimal interpolation objective analysis of the Argo data, improvements are made to the em- pirical formula of a background error covariance matrix widely used in data assimilation and objective anal- ysis systems. Specifically, an estimation of correlation scales that can improve effectively the accuracy of Ar- go objective analysis has been developed. This method can automatically adapt to the gradient change of a variable and is referred to as ＂gradient-dependent correlation scale method＂. Its effect on the Argo objective analysis is verified theoretically with Gaussian pulse and spectrum analysis. The results of one-dimensional simulation experiment show that the gradient-dependent correlation scales can improve the adaptability of the objective analysis system, making it possible for the analysis scheme to fully absorb the shortwave information of observation in areas with larger oceanographic gradients. The new scheme is applied to the Argo data obiective analysis system in the Pacific Ocean. The results are obviously improved.展开更多
An ocean reanalysis system for the joining area of Asia and Indian-Pacific Ocean (AIPO) has been developed and is currently delivering reanalysis data sets for study on the air-sea interaction over AIPO and its climat...An ocean reanalysis system for the joining area of Asia and Indian-Pacific Ocean (AIPO) has been developed and is currently delivering reanalysis data sets for study on the air-sea interaction over AIPO and its climate variation over China in the inter-annual time scale.This system consists of a nested ocean model forced by atmospheric reanalysis,an ensemble-based multivariate ocean data assimilation system and various ocean observations.The following report describes the main components of the data assimilation system in detail.The system adopts an ensemble optimal interpolation scheme that uses a seasonal update from a free running model to estimate the background error covariance matrix.In view of the systematic biases in some observation systems,some treatments were performed on the observations before the assimilation.A coarse resolution reanalysis dataset from the system is preliminarily evaluated to demonstrate the performance of the system for the period 1992 to 2006 by comparing this dataset with other observations or reanalysis data.展开更多
The development and application of a regional ocean data assimilation system are among the aims of the Global Ocean Data Assimilation Experiment. The ocean data assimilation system in the regions including the Indian ...The development and application of a regional ocean data assimilation system are among the aims of the Global Ocean Data Assimilation Experiment. The ocean data assimilation system in the regions including the Indian and West Pacific oceans is an endeavor motivated by this goal. In this study, we describe the system in detail. Moreover, the reanalysis in the joint area of Asia, the Indian Ocean, and the western Pacific Ocean （hereafter AIPOcean） constructed using multi-year model integration with data assimilation is used to test the performance of this system. The ocean model is an eddy-resolving, hybrid coordinate ocean model. Various types of observations including in-situ temperature and salinity profiles （mechanical bathythermograph, expendable bathythermograph, Array for Real-time Geostrophic Oceanography, Tropical Atmosphere Ocean Array, conductivity-temperature-depth, station data）, remotely-sensed sea surface temperature, and altimetry sea level anomalies, are assimilated into the reanalysis via the ensemble optimal interpolation method. An ensemble of model states sampled from a long-term integration is allowed to change with season, rather than remaining stationary. The estimated background error covariance matrix may reasonably reflect the seasonality and anisotropy. We evaluate the performance of AIPOcean during the period 1993-2006 by comparisons with independent observations, and some reanalysis products. We show that AIPOcean reduces the errors of subsurface temperature and salinity, and reproduces mesoscale eddies. In contrast to ECCO and SODA products, AIPOcean captures the interannual variability and linear trend of sea level anomalies very well. AIPOcean also shows a good consistency with tide gauges.展开更多
In order to improve the efficiency of the Ocean Variational Assimilation System （OVALS）, which has been widely used in various applications, an improved OVALS （OVALS2） is developed based on the recursive filter ...In order to improve the efficiency of the Ocean Variational Assimilation System （OVALS）, which has been widely used in various applications, an improved OVALS （OVALS2） is developed based on the recursive filter （RF） algorithm. The first advantage of OVALS2 is that memory storage can be substantially reduced in practice because it implicitly computes the background error covariance matrix; the second advantage is that there is no inversion of the background error covariance by preconditioning the control variable. For comparing the effectiveness between OVALS2 and OVALS, a set of experiments was implemented by assimilating expendable bathythermograph （XBT） and ARGO data into the Tropical Pacific circulation model. The results show that the efficiency of OVALS2 is much higher than that of OVALS. The computational time and the computer storage in the assimilation process were reduced by 83% and 77%, respectively. Additionally, the corresponding results produced by the RF are almost as good as those obtained by OVALS. These results prove that OVALS2 is suitable for operational numerical oceanic forecasting.展开更多
A data assimilation scheme used in the updated Ocean three-dimensional Variational Assimila- tion System （OVALS）, OVALS2, is described. Based on a recursive filter （RF） to estimate the background error covariance ...A data assimilation scheme used in the updated Ocean three-dimensional Variational Assimila- tion System （OVALS）, OVALS2, is described. Based on a recursive filter （RF） to estimate the background error covariance （BEC） over a predetermined scale, this new analysis system can be implemented with anisotropic and isotropic BECs. Similarities and differences of these two BEC schemes are briefly discussed and their impacts on the model simulation are also investigated. An idealized experiment demonstrates the ability of the updated analysis system to construct different BECs. Furthermore, a set of three years experiments is implemented by assimilating expendable bathythermograph （XBT） and ARGO data into a Tropical Pacific circulation model. The TAO and WOA01 data are used to validate the assimilation results. The results show that the model simu- lations are substantially improved by OVALS2. The inter-comparison of isotropic and anisotropic BEC shows that the corresponding temperature and salinity produced by the anisotropic BEC are almost as good as those obtained by the isotropic one. Moreover, the result of anisotropic RF is slightly closer to WOA01 and TAO than that of isotropic RF in some special area （e.g. the cold tongue area in the Tropic Pacific）.展开更多
Despite a specific data assimilation method,data assimilation(DA)in general can be decomposed into components of the prior information,observation forward operator that is given by the observation type,observation err...Despite a specific data assimilation method,data assimilation(DA)in general can be decomposed into components of the prior information,observation forward operator that is given by the observation type,observation error covariances,and background error covariances.In a classic Lorenz model,the influences of the DA components on the initial conditions(ICs)and subsequent forecasts are systematically investigated,which could provide a theoretical basis for the design of DA for different scales of interests.The forecast errors undergo three typical stages:a slow growth stage from 0 h to 5 d,a fast growth stage from 5 d to around 15 d with significantly different error growth rates for ensemble and deterministic forecasts,and a saturation stage after 15 d.Assimilation strategies that provide more accurate ICs can improve the predictability.Cycling assimilation is superior to offline assimilation,and a flow-dependent background error covariance matrix(Pf)provides better analyses than a static background error covariance matrix(B)for instantaneous observations and frequent time-averaged observations;but the opposite is true for infrequent time-averaged observations,since cycling simulation cannot construct informative priors when the model lacks predictive skills and the flow-dependent Pf cannot effectively extract information from low-informative observations as the static B.Instantaneous observations contain more information than time-averaged observations,thus the former is preferred,especially for infrequent observing systems.Moreover,ensemble forecasts have advantages over deterministic forecasts,and the advantages are enlarged with less informative observations and lower predictive-skill model priors.展开更多
Ocean observations are inherently characterized by irregular temporal and spatial distributions,as well as heterogeneous spatial resolutions and error characteristics arising from the use of diverse observational plat...Ocean observations are inherently characterized by irregular temporal and spatial distributions,as well as heterogeneous spatial resolutions and error characteristics arising from the use of diverse observational platforms and techniques.To enable their application across a broad range of scientific and practical problems,it is essential to map these heterogeneous datasets into temporally and spatially consistent gridded products.Optimal Interpolation remains the most widely adopted algorithm for the mapping of oceanographic data.Two principal implementations of the optimal interpolation algorithm are commonly employed.The first,known as the basic optimal interpolation,is derived from the theory of optimal estimation and involves computationally intensive matrix operations,posing significant challenges when applied to high-dimensional problems.The second,referred to as the point-wise optimal interpolation,reduces computational complexity through point-wise estimation,thereby circumventing high-dimensional operations;however,this approach results in a substantially higher overall computational cost.In this study,a novel optimal interpolation algorithm is proposed that utilizes the Kronecker product to approximate the background error covariance matrix.This formulation enables the decomposition of high-dimensional matrix operations into smaller,computationally tractable sub-problems,thereby improving the scalability of optimal interpolation for large spatial domains with dense observational coverage.Building upon this framework,a multi-scale optimal interpolation method is further developed to enhance the integration of observational datasets with widely varying spatial resolutions,thereby improving the accuracy and applicability of the resulting gridded products.展开更多
Accurate estimation of the background error covariance matrix denoted as B remains a critical challenge in numerical weather prediction(NWP),directly influencing data assimilation(DA)performance and forecast accuracy....Accurate estimation of the background error covariance matrix denoted as B remains a critical challenge in numerical weather prediction(NWP),directly influencing data assimilation(DA)performance and forecast accuracy.Although hybrid ensemble–variational(En Var)methods combine static and flow-dependent matrices to improve assimilation,their effectiveness is constrained by empirically fixed weights.To address this limitation,we propose DRL-En Var,an adaptive hybrid En Var DA method enhanced with deep reinforcement learning.DRL-En Var integrates deep learning(DL)components,including a novel cyclic convolution module to extract abstract features from data,and employs reinforcement learning(RL)to dynamically optimize hybrid weighting strategies.The system adaptively combines multiple ensemble-based flow-dependent matrices with one or more static matrices to construct a time-varying hybrid matrix B that better reflects real-time background errors.Experimental results demonstrate that DRL-En Var performs better than the traditional ensemble Kalman filter(En KF)and hybrid covariance DA(HCDA)methods,especially under sparse observations or transitional changes in state variables.It achieves competitive or superior assimilation accuracy with lower computational cost,and can be flexibly integrated into both three-dimensional variational assimilation(3DVar)and four-dimensional variational assimilation(4DVar)frameworks.Overall,DRL-En Var offers a novel and efficient approach to adaptive DA,particularly valuable for improving forecast skill during transitional weather regimes.展开更多
基金sponsored by the National Natural Science Foundation of China[grant number U2442218]。
文摘Land surface temperature(LST)is the key variable in land-atmosphere interaction,having an important impact on weather and climate forecasting.However,achieving consistent analysis of LST and the atmosphere in assimilation is quite challenging.This is because there is limited knowledge about the cross-component background error covariance(BEC)between LST and atmospheric state variables.This study aims to clarify whether there is a relationship between the error of LST and atmospheric variables,and whether this relationship varies spatially and temporally.To this end,the BEC coupled with atmospheric variables and LST was constructed(LST-BEC),and its characteristics were analyzed based on the 2023 mei-yu season.The general characteristics of LST-BEC show that the LST is mainly correlated with the atmospheric temperature and the correlation decreases gradually with a rise in atmospheric height,and the error standard deviation of the LST is noticeably larger than that of the low-level atmospheric temperature.The spatiotemporal characteristics of LST-BEC on the heavy-rain day and light-rain day show that the error correlation and error standard deviation of LST and low-level atmospheric temperature and humidity are closely related to the weather background,and also have obvious diurnal variations.These results provide valuable information for strongly coupled land-atmosphere assimilation.
基金National Natural Science Foundation of China (40775064)
文摘Background error covariance plays an important role in any variational data assimilation system, because it determines how information from observations is spread in model space and between different model variables. In this paper, the use of orthogonal wavelets in representation of background error covariance over a limited area is studied. Based on the WRF model and its 3D-VAR system, an algorithm using orthogonal wavelets to model background error covariance is developed. Because each wavelet function contains information on both position and scale, using a diagonal correlation matrix in wavelet space gives the possibility to represent some anisotropic and inhomogeneous characteristics of background error covariance. The experiments show that local correlation functions are better modeled than spectral methods. The formulation of wavelet background error covariance is tested with the typhoon Kaemi (2006). The results of experiments indicate that the subsequent forecasts of typhoon Kaemi’s track and intensity are significantly improved by the new method.
基金Supported by the National Natural Science Foundation of China (Nos.40806011,U1133001)the Open Fund of the Key Laboratory of Ocean Circulation and Waves,Chinese Academy of Sciences(No. KLOCAW0806)
文摘The quality of background error statistics is one of the key components for successful assimilation of observations in a numerical model.The background error covariance(BEC) of ocean waves is generally estimated under an assumption that it is stationary over a period of time and uniform over a domain.However,error statistics are in fact functions of the physical processes governing the meteorological situation and vary with the wave condition.In this paper,we simulated the BEC of the significant wave height(SWH) employing Monte Carlo methods.An interesting result is that the BEC varies consistently with the mean wave direction(MWD).In the model domain,the BEC of the SWH decreases significantly when the MWD changes abruptly.A new BEC model of the SWH based on the correlation between the BEC and MWD was then developed.A case study of regional data assimilation was performed,where the SWH observations of buoy 22001 were used to assess the SWH hindcast.The results show that the new BEC model benefits wave prediction and allows reasonable approximations of anisotropy and inhomogeneous errors.
基金National Key Research and Development Program of China(Grant No.2017YFC1502102)National Natural Science Foundation of China(Grant No.42075148)+1 种基金Graduate Research and Innovation Projects of Jiangsu Province(Grant No.KYCX20_0910)the High-Performance Computing Center of Nanjing University of Information Science and Technology(NUIST).
文摘Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor control variables are necessary.Common data assimilation systems theoretically require that the probability density functions(PDFs)of analysis,background,and observation errors should satisfy the Gaussian unbiased assumptions.In this study,a Gaussian transform method is proposed to transform hydrometeors to more Gaussian variables,which is modified from the Softmax function and renamed as Quasi-Softmax transform.The Quasi-Softmax transform method then is compared to the original hydrometeor mixing ratios and their logarithmic transform and Softmax transform.The spatial distribution,the non-Gaussian nature of the background errors,and the characteristics of the background errors of hydrometeors in each method are studied.Compared to the logarithmic and Softmax transform,the Quasi-Softmax method keeps the vertical distribution of the original hydrometeor mixing ratios to the greatest extent.The results of the D′Agostino test show that the hydrometeors transformed by the Quasi-Softmax method are more Gaussian when compared to the other methods.The Gaussian transform has been added to the control variable transform to estimate the background error covariances.Results show that the characteristics of the hydrometeor background errors are reasonable for the Quasi-Softmax method.The transformed hydrometeors using the Quasi-Softmax transform meet the Gaussian unbiased assumptions of the data assimilation system,and are promising control variables for data assimilation systems.
基金funded by the Korea Meteorological Administration Research and Development Program under Grant RACS 2010-2016supported by the Brain Korea 21 project of the Ministry of Education and Human Resources Development of the Korean government
文摘Satellite data obtained over synoptic data-sparse regions such as an ocean contribute toward improving the quality of the initial state of limited-area models. Background error covariances are crucial to the proper distribution of satellite-observed information in variational data assimilation. In the NMC （National Meteorological Center） method, background error covariances are underestimated over data-sparse regions such as an ocean because of small differences between different forecast times. Thus, it is necessary to reconstruct and tune the background error covariances so as to maximize the usefulness of the satellite data for the initial state of limited-area models, especially over an ocean where there is a lack of conventional data. In this study, we attempted to estimate background error covariances so as to provide adequate error statistics for data-sparse regions by using ensemble forecasts of optimal perturbations using bred vectors. The background error covariances estimated by the ensemble method reduced the overestimation of error amplitude obtained by the NMC method. By employing an appropriate horizontal length scale to exclude spurious correlations, the ensemble method produced better results than the NMC method in the assimilation of retrieved satellite data. Because the ensemble method distributes observed information over a limited local area, it would be more useful in the analysis of high-resolution satellite data. Accordingly, the performance of forecast models can be improved over the area where the satellite data are assimilated.
基金This work was jointly sponsored by the National Natural Science Foundation of China[grant number 42075148]the Outreach Projects of the State Key Laboratory of Severe Weather[grant number 2021LASWA08]+1 种基金the Outreach Projects of the Key Laboratory of Meteorological Disaster[grant number KLME202209]supported by the High-Performance Computing Center of Nanjing University of Information Science and Technology(NUIST).
文摘Background error covariance(BEC)plays an essential role in variational data assimilation.Most variational data assimilation systems still use static BEC.Actually,the characteristics of BEC vary with season,day,and even hour of the background.National Meteorological Center-based diurnally varying BECs had been proposed,but the diurnal variation characteristics were gained by climatic samples.Ensemble methods can obtain the background error characteristics that suit the samples in the current moment.Therefore,to gain more reasonable diurnally varying BECs,in this study,ensemble-based diurnally varying BECs are generated and the diurnal variation characteristics are discussed.Their impacts are then evaluated by cycling data assimilation and forecasting experiments for a week based on the operational China Meteorological Administration-Beijing system.Clear diurnal variation in the standard deviation of ensemble forecasts and ensemble-based BECs can be identified,consistent with the diurnal variation characteristics of the atmosphere.The results of one-week cycling data assimilation and forecasting show that the application of diurnally varying BECs reduces the RMSEs in the analysis and 6-h forecast.Detailed analysis of a convective rainfall case shows that the distribution of the accumulated precipitation forecast using the diurnally varying BECs is closer to the observation than using the static BEC.Besides,the cycle-averaged precipitation scores in all magnitudes are improved,especially for the heavy precipitation,indicating the potential of using diurnally varying BEC in operational applications.
基金National Natural Science Foundation of China through Grants(41461164008,41130964)National Key Project for Basic Research(973 Project)(2015452803)+1 种基金Science and Technology Planning Project for Guangdong Province(2012A061400012)China Meteorological Administration(GYHY201406009)
文摘In the previous study, the influences of introducing larger- and smaller-scale errors on the background error covariances estimated at the given scales were investigated, respectively. This study used the eovariances obtained in the previous study in the data assimilation and model forecast system based on three-dimensional variational method and the Weather Research and Forecasting model. In this study, analyses and forecasts from this system with different covariances for a period of one month were compared, and the causes for differing results were presented. The varia- tions of analysis increments with different-scale errors are consistent with those of variances and correlations of back- ground errors that were reported in the previous paper. In particular, the introduction of smaller-scale errors leads to greater amplitudes in analysis increments for medium-scale wind at the heights of both high- and low-level jets. Tem- perature and humidity analysis increments are greater at the corresponding scales at the middle- and upper-levels. These analysis increments could improve the intensity of the jet-convection system that includes jets at different levels and the coupling between them that is associated with latent heat release. These changes in analyses will contribute to more ac- curate wind and temperature forecasts in the corresponding areas. When smaller-scale errors are included, humidity analysis increments are significantly enhanced at large scales and lower levels, to moisten southern analyses. Thus, dry bias can be corrected, which will improve humidity forecasts. Moreover, the inclusion of larger- （smaller-） scale errors will be beneficial for the accuracy of forecasts of heavy （light） precipitation at large （small） scales because of the ampli- fication （diminution） of the intensity and area in precipitation forecasts.
基金The Marine Public Welfare Special Funds,the State Oceanic Administration of China under contract No.200705022the Technology Special Basic Work,the Ministry of Science and Technology under contract No.2012FY112300the Basic Scientific Research Special Funds of the Second Institute of Oceanography,the State Oceanic Administration of China under contract No.IT0904
文摘Based on the optimal interpolation objective analysis of the Argo data, improvements are made to the em- pirical formula of a background error covariance matrix widely used in data assimilation and objective anal- ysis systems. Specifically, an estimation of correlation scales that can improve effectively the accuracy of Ar- go objective analysis has been developed. This method can automatically adapt to the gradient change of a variable and is referred to as ＂gradient-dependent correlation scale method＂. Its effect on the Argo objective analysis is verified theoretically with Gaussian pulse and spectrum analysis. The results of one-dimensional simulation experiment show that the gradient-dependent correlation scales can improve the adaptability of the objective analysis system, making it possible for the analysis scheme to fully absorb the shortwave information of observation in areas with larger oceanographic gradients. The new scheme is applied to the Argo data obiective analysis system in the Pacific Ocean. The results are obviously improved.
基金supported by the Chinese Academy of Sciences (Grant No. KZCX2-YW-202)the 973 Pro-gram (Grant No. 2006CB403606),the 863 Program (Grant No.2009AA12Z138)the National Natural Science Foundation of China (Grant Nos. 40606008,40437017,and 40221503)
文摘An ocean reanalysis system for the joining area of Asia and Indian-Pacific Ocean (AIPO) has been developed and is currently delivering reanalysis data sets for study on the air-sea interaction over AIPO and its climate variation over China in the inter-annual time scale.This system consists of a nested ocean model forced by atmospheric reanalysis,an ensemble-based multivariate ocean data assimilation system and various ocean observations.The following report describes the main components of the data assimilation system in detail.The system adopts an ensemble optimal interpolation scheme that uses a seasonal update from a free running model to estimate the background error covariance matrix.In view of the systematic biases in some observation systems,some treatments were performed on the observations before the assimilation.A coarse resolution reanalysis dataset from the system is preliminarily evaluated to demonstrate the performance of the system for the period 1992 to 2006 by comparing this dataset with other observations or reanalysis data.
基金supported by the 973 Program (Grant No.2010CB950401)the Chinese Academy of Sciences’ Project"Western Pacific Ocean System:Structure,Dynamics and Consequences"(Grant No.XDA11010405)the National Natural Science Foundation of China (Grant No.41176015)
文摘The development and application of a regional ocean data assimilation system are among the aims of the Global Ocean Data Assimilation Experiment. The ocean data assimilation system in the regions including the Indian and West Pacific oceans is an endeavor motivated by this goal. In this study, we describe the system in detail. Moreover, the reanalysis in the joint area of Asia, the Indian Ocean, and the western Pacific Ocean （hereafter AIPOcean） constructed using multi-year model integration with data assimilation is used to test the performance of this system. The ocean model is an eddy-resolving, hybrid coordinate ocean model. Various types of observations including in-situ temperature and salinity profiles （mechanical bathythermograph, expendable bathythermograph, Array for Real-time Geostrophic Oceanography, Tropical Atmosphere Ocean Array, conductivity-temperature-depth, station data）, remotely-sensed sea surface temperature, and altimetry sea level anomalies, are assimilated into the reanalysis via the ensemble optimal interpolation method. An ensemble of model states sampled from a long-term integration is allowed to change with season, rather than remaining stationary. The estimated background error covariance matrix may reasonably reflect the seasonality and anisotropy. We evaluate the performance of AIPOcean during the period 1993-2006 by comparisons with independent observations, and some reanalysis products. We show that AIPOcean reduces the errors of subsurface temperature and salinity, and reproduces mesoscale eddies. In contrast to ECCO and SODA products, AIPOcean captures the interannual variability and linear trend of sea level anomalies very well. AIPOcean also shows a good consistency with tide gauges.
基金supported by the Chinese Academy of Science(Contract No. KZCX2-YW-202)the 973 Pro-gram (Grant No. 2006CB403606)the National Natural Science Foundation of China (Grant Nos. 40606008,40776011)
文摘In order to improve the efficiency of the Ocean Variational Assimilation System （OVALS）, which has been widely used in various applications, an improved OVALS （OVALS2） is developed based on the recursive filter （RF） algorithm. The first advantage of OVALS2 is that memory storage can be substantially reduced in practice because it implicitly computes the background error covariance matrix; the second advantage is that there is no inversion of the background error covariance by preconditioning the control variable. For comparing the effectiveness between OVALS2 and OVALS, a set of experiments was implemented by assimilating expendable bathythermograph （XBT） and ARGO data into the Tropical Pacific circulation model. The results show that the efficiency of OVALS2 is much higher than that of OVALS. The computational time and the computer storage in the assimilation process were reduced by 83% and 77%, respectively. Additionally, the corresponding results produced by the RF are almost as good as those obtained by OVALS. These results prove that OVALS2 is suitable for operational numerical oceanic forecasting.
基金Major National Scientific Research Project on Global Change under contract No. 2010CB951901the National Science Foundation of China under contract No. 40821092+2 种基金Special Fund for Public Welfare Industry (Meteorology) under contract No.GYHY200906018supported by the Natural Science Foundation of China under Contract No. 40805033Key Projects in the National Science & Technology Pillar Program during the Eleventh Five-Year Plan Period under Contract No. 2006BAC03B03
文摘A data assimilation scheme used in the updated Ocean three-dimensional Variational Assimila- tion System （OVALS）, OVALS2, is described. Based on a recursive filter （RF） to estimate the background error covariance （BEC） over a predetermined scale, this new analysis system can be implemented with anisotropic and isotropic BECs. Similarities and differences of these two BEC schemes are briefly discussed and their impacts on the model simulation are also investigated. An idealized experiment demonstrates the ability of the updated analysis system to construct different BECs. Furthermore, a set of three years experiments is implemented by assimilating expendable bathythermograph （XBT） and ARGO data into a Tropical Pacific circulation model. The TAO and WOA01 data are used to validate the assimilation results. The results show that the model simu- lations are substantially improved by OVALS2. The inter-comparison of isotropic and anisotropic BEC shows that the corresponding temperature and salinity produced by the anisotropic BEC are almost as good as those obtained by the isotropic one. Moreover, the result of anisotropic RF is slightly closer to WOA01 and TAO than that of isotropic RF in some special area （e.g. the cold tongue area in the Tropic Pacific）.
基金This work was supported by the National Natural Science Foundation of China(Grant Nos.42192553,41922036&41775057)the Frontiers Science Center for Critical Earth Material Cycling Fund(Grant No.JBGS2102)the Fundamental Research Funds for the Central Universities(Grant No.0209-14380097).
文摘Despite a specific data assimilation method,data assimilation(DA)in general can be decomposed into components of the prior information,observation forward operator that is given by the observation type,observation error covariances,and background error covariances.In a classic Lorenz model,the influences of the DA components on the initial conditions(ICs)and subsequent forecasts are systematically investigated,which could provide a theoretical basis for the design of DA for different scales of interests.The forecast errors undergo three typical stages:a slow growth stage from 0 h to 5 d,a fast growth stage from 5 d to around 15 d with significantly different error growth rates for ensemble and deterministic forecasts,and a saturation stage after 15 d.Assimilation strategies that provide more accurate ICs can improve the predictability.Cycling assimilation is superior to offline assimilation,and a flow-dependent background error covariance matrix(Pf)provides better analyses than a static background error covariance matrix(B)for instantaneous observations and frequent time-averaged observations;but the opposite is true for infrequent time-averaged observations,since cycling simulation cannot construct informative priors when the model lacks predictive skills and the flow-dependent Pf cannot effectively extract information from low-informative observations as the static B.Instantaneous observations contain more information than time-averaged observations,thus the former is preferred,especially for infrequent observing systems.Moreover,ensemble forecasts have advantages over deterministic forecasts,and the advantages are enlarged with less informative observations and lower predictive-skill model priors.
基金The National Key Research and Development Program of China under contract No.2022YFF0801404.
文摘Ocean observations are inherently characterized by irregular temporal and spatial distributions,as well as heterogeneous spatial resolutions and error characteristics arising from the use of diverse observational platforms and techniques.To enable their application across a broad range of scientific and practical problems,it is essential to map these heterogeneous datasets into temporally and spatially consistent gridded products.Optimal Interpolation remains the most widely adopted algorithm for the mapping of oceanographic data.Two principal implementations of the optimal interpolation algorithm are commonly employed.The first,known as the basic optimal interpolation,is derived from the theory of optimal estimation and involves computationally intensive matrix operations,posing significant challenges when applied to high-dimensional problems.The second,referred to as the point-wise optimal interpolation,reduces computational complexity through point-wise estimation,thereby circumventing high-dimensional operations;however,this approach results in a substantially higher overall computational cost.In this study,a novel optimal interpolation algorithm is proposed that utilizes the Kronecker product to approximate the background error covariance matrix.This formulation enables the decomposition of high-dimensional matrix operations into smaller,computationally tractable sub-problems,thereby improving the scalability of optimal interpolation for large spatial domains with dense observational coverage.Building upon this framework,a multi-scale optimal interpolation method is further developed to enhance the integration of observational datasets with widely varying spatial resolutions,thereby improving the accuracy and applicability of the resulting gridded products.
基金Project supported by the National Key R&D Program of China(No.2022YFB3207304)the National Natural Science Foundation of China(No.42205161)the Natural Science Foundation of Hunan Province,China(No.2023JJ30630)。
文摘Accurate estimation of the background error covariance matrix denoted as B remains a critical challenge in numerical weather prediction(NWP),directly influencing data assimilation(DA)performance and forecast accuracy.Although hybrid ensemble–variational(En Var)methods combine static and flow-dependent matrices to improve assimilation,their effectiveness is constrained by empirically fixed weights.To address this limitation,we propose DRL-En Var,an adaptive hybrid En Var DA method enhanced with deep reinforcement learning.DRL-En Var integrates deep learning(DL)components,including a novel cyclic convolution module to extract abstract features from data,and employs reinforcement learning(RL)to dynamically optimize hybrid weighting strategies.The system adaptively combines multiple ensemble-based flow-dependent matrices with one or more static matrices to construct a time-varying hybrid matrix B that better reflects real-time background errors.Experimental results demonstrate that DRL-En Var performs better than the traditional ensemble Kalman filter(En KF)and hybrid covariance DA(HCDA)methods,especially under sparse observations or transitional changes in state variables.It achieves competitive or superior assimilation accuracy with lower computational cost,and can be flexibly integrated into both three-dimensional variational assimilation(3DVar)and four-dimensional variational assimilation(4DVar)frameworks.Overall,DRL-En Var offers a novel and efficient approach to adaptive DA,particularly valuable for improving forecast skill during transitional weather regimes.
