Conditional nonlinear optimal perturbation(CNOP) is an extension of the linear singular vector technique in the nonlinear regime.It represents the initial perturbation that is subjected to a given physical constraint,...Conditional nonlinear optimal perturbation(CNOP) is an extension of the linear singular vector technique in the nonlinear regime.It represents the initial perturbation that is subjected to a given physical constraint,and results in the largest nonlinear evolution at the prediction time.CNOP-type errors play an important role in the predictability of weather and climate.Generally,when calculating CNOP in a complicated numerical model,we need the gradient of the objective function with respect to the initial perturbations to provide the descent direction for searching the phase space.The adjoint technique is widely used to calculate the gradient of the objective function.However,it is difficult and cumbersome to construct the adjoint model of a complicated numerical model,which imposes a limitation on the application of CNOP.Based on previous research,this study proposes a new ensemble projection algorithm based on singular vector decomposition(SVD).The new algorithm avoids the localization procedure of previous ensemble projection algorithms,and overcomes the uncertainty caused by choosing the localization radius empirically.The new algorithm is applied to calculate the CNOP in an intermediate forecasting model.The results show that the CNOP obtained by the new ensemble-based algorithm can effectively approximate that calculated by the adjoint algorithm,and retains the general spatial characteristics of the latter.Hence,the new SVD-based ensemble projection algorithm proposed in this study is an effective method of approximating the CNOP.展开更多
Based on 20 models from phase 6 of the Coupled Model Intercomparison Project(CMIP6),this article explored possible reasons for differences in simulation biases and projected changes in precipitation in northern China ...Based on 20 models from phase 6 of the Coupled Model Intercomparison Project(CMIP6),this article explored possible reasons for differences in simulation biases and projected changes in precipitation in northern China among the allmodel ensemble(AMME),“highest-ranked”model ensemble(BMME),and“lowest-ranked”model ensemble(WMME),from the perspective of atmospheric circulations and moisture budgets.The results show that the BMME and AMME reproduce the East Asian winter circulations better than the WMME.Compared with the AMME and WMME,the BMME reduces the overestimation of evaporation,thereby improving the simulation of winter precipitation.The three ensemble simulated biases for the East Asian summer circulations are generally similar,characterized by a stronger zonal pressure gradient between the mid-latitudes of the North Pacific and East Asia and a northward displacement of the East Asian westerly jet.However,the simulated vertical moisture advection is improved in the BMME,contributing to the slightly higher performance of the BMME than the AMME and WMME on summer precipitation in North and Northeast China.Compared to the AMME and WMME,the BMME projects larger increases in precipitation in northern China during both seasons by the end of the 21st century under the Shared Socioeconomic Pathway 5-8.5(SSP5-8.5).One of the reasons is that the increase in evaporation projected by the BMME is larger.The projection of a greater dynamic contribution by the BMME also plays a role.In addition,larger changes in the nonlinear components in the BMME projection contribute to a larger increase in winter precipitation in northern China.展开更多
We present new variants of Estimation of Distribution Algorithms (EDA) for large-scale continuous optimisation that extend and enhance a recently proposed random projection (RP) ensemble based approach. The main novel...We present new variants of Estimation of Distribution Algorithms (EDA) for large-scale continuous optimisation that extend and enhance a recently proposed random projection (RP) ensemble based approach. The main novelty here is to depart from the theory of RPs that require (sub-)Gaussian random matrices for norm-preservation, and instead for the purposes of high-dimensional search we propose to employ random matrices with independent and identically distributed entries drawn from a t-distribution. We analytically show that the implicitly resulting high-dimensional covariance of the search distribution is enlarged as a result. Moreover, the extent of this enlargement is controlled by a single parameter, the degree of freedom. For this reason, in the context of optimisation, such heavy tailed random matrices turn out to be preferable over the previously employed (sub-)Gaussians. Based on this observation, we then propose novel covariance adaptation schemes that are able to adapt the degree of freedom parameter during the search, and give rise to a flexible approach to balance exploration versus exploitation. We perform a thorough experimental study on high-dimensional benchmark functions, and provide statistical analyses that demonstrate the state-of-the-art performance of our approach when compared with existing alternatives in problems with 1000 search variables.展开更多
Changes in temperature and precipitation have a profound effect on the ecological environment and socioeconomic systems.In this study,we focus on the major Belt and Road Initiative(BRI)regions and develop a dataset of...Changes in temperature and precipitation have a profound effect on the ecological environment and socioeconomic systems.In this study,we focus on the major Belt and Road Initiative(BRI)regions and develop a dataset of temperature and precipitation at global temperature rise targets of 1.5°C,2°C,and 3°C above pre-industrial levels under the Representative Concentration Pathway(RCP)8.5 emission scenario using 4 downscaled global model datasets data at a fine spatial resolution of 0.0449147848°(~5 km)globally from EnviDat.The temperature variables include the daily maximum(Tmax),minimum(Tmin)and average(Tmp)surface air temperatures,and the diurnal temperature range(DTR).We first evaluate the performance of the downscaled model data using CRU-observed gridded data for the historical period 1986-2005.The results indicate that the downscaled model data can generally reproduce the pattern characteristics of temperature and precipitation variations well over the major BRI regions for 1986-2005.Furthermore,we project temperature and precipitation variations over the major BRI regions at global temperature rise targets of 1.5°C,2°C,and 3°C under the RCP8.5 emission scenario based on the dataset by adopting the multiple-model ensemble mean.Our dataset contributes to understanding detailed the characteristics of climate change over the major BRI regions,and provides data fundamental for adopting appropriate strategies and options to reduce or avoid disadvantaged consequences associated with climate change over the major BRI regions.The dataset is available at https://doi.org/10.57760/sciencedb.01850.展开更多
基金jointly sponsored by the National Natural Science Foundation of China(Grant Nos.41176013,41230420 and 41006007)
文摘Conditional nonlinear optimal perturbation(CNOP) is an extension of the linear singular vector technique in the nonlinear regime.It represents the initial perturbation that is subjected to a given physical constraint,and results in the largest nonlinear evolution at the prediction time.CNOP-type errors play an important role in the predictability of weather and climate.Generally,when calculating CNOP in a complicated numerical model,we need the gradient of the objective function with respect to the initial perturbations to provide the descent direction for searching the phase space.The adjoint technique is widely used to calculate the gradient of the objective function.However,it is difficult and cumbersome to construct the adjoint model of a complicated numerical model,which imposes a limitation on the application of CNOP.Based on previous research,this study proposes a new ensemble projection algorithm based on singular vector decomposition(SVD).The new algorithm avoids the localization procedure of previous ensemble projection algorithms,and overcomes the uncertainty caused by choosing the localization radius empirically.The new algorithm is applied to calculate the CNOP in an intermediate forecasting model.The results show that the CNOP obtained by the new ensemble-based algorithm can effectively approximate that calculated by the adjoint algorithm,and retains the general spatial characteristics of the latter.Hence,the new SVD-based ensemble projection algorithm proposed in this study is an effective method of approximating the CNOP.
基金jointly supported by the National Natural Science Foundation of China(Grant No.41991285)the National Key Research and Development Program of China(2017YFA0605004)the Program for Distinguished Professors of Jiangsu。
文摘Based on 20 models from phase 6 of the Coupled Model Intercomparison Project(CMIP6),this article explored possible reasons for differences in simulation biases and projected changes in precipitation in northern China among the allmodel ensemble(AMME),“highest-ranked”model ensemble(BMME),and“lowest-ranked”model ensemble(WMME),from the perspective of atmospheric circulations and moisture budgets.The results show that the BMME and AMME reproduce the East Asian winter circulations better than the WMME.Compared with the AMME and WMME,the BMME reduces the overestimation of evaporation,thereby improving the simulation of winter precipitation.The three ensemble simulated biases for the East Asian summer circulations are generally similar,characterized by a stronger zonal pressure gradient between the mid-latitudes of the North Pacific and East Asia and a northward displacement of the East Asian westerly jet.However,the simulated vertical moisture advection is improved in the BMME,contributing to the slightly higher performance of the BMME than the AMME and WMME on summer precipitation in North and Northeast China.Compared to the AMME and WMME,the BMME projects larger increases in precipitation in northern China during both seasons by the end of the 21st century under the Shared Socioeconomic Pathway 5-8.5(SSP5-8.5).One of the reasons is that the increase in evaporation projected by the BMME is larger.The projection of a greater dynamic contribution by the BMME also plays a role.In addition,larger changes in the nonlinear components in the BMME projection contribute to a larger increase in winter precipitation in northern China.
基金partly funded by a Ph.D.scholarship from the Islamic Development Bankfunded by the Engineering and Physical Sciences Research Council of UK under Fellowship Grant EP/P004245/1.
文摘We present new variants of Estimation of Distribution Algorithms (EDA) for large-scale continuous optimisation that extend and enhance a recently proposed random projection (RP) ensemble based approach. The main novelty here is to depart from the theory of RPs that require (sub-)Gaussian random matrices for norm-preservation, and instead for the purposes of high-dimensional search we propose to employ random matrices with independent and identically distributed entries drawn from a t-distribution. We analytically show that the implicitly resulting high-dimensional covariance of the search distribution is enlarged as a result. Moreover, the extent of this enlargement is controlled by a single parameter, the degree of freedom. For this reason, in the context of optimisation, such heavy tailed random matrices turn out to be preferable over the previously employed (sub-)Gaussians. Based on this observation, we then propose novel covariance adaptation schemes that are able to adapt the degree of freedom parameter during the search, and give rise to a flexible approach to balance exploration versus exploitation. We perform a thorough experimental study on high-dimensional benchmark functions, and provide statistical analyses that demonstrate the state-of-the-art performance of our approach when compared with existing alternatives in problems with 1000 search variables.
基金the National Key R&D Program of China(2018YFA0606501,2017YFA0603601).
文摘Changes in temperature and precipitation have a profound effect on the ecological environment and socioeconomic systems.In this study,we focus on the major Belt and Road Initiative(BRI)regions and develop a dataset of temperature and precipitation at global temperature rise targets of 1.5°C,2°C,and 3°C above pre-industrial levels under the Representative Concentration Pathway(RCP)8.5 emission scenario using 4 downscaled global model datasets data at a fine spatial resolution of 0.0449147848°(~5 km)globally from EnviDat.The temperature variables include the daily maximum(Tmax),minimum(Tmin)and average(Tmp)surface air temperatures,and the diurnal temperature range(DTR).We first evaluate the performance of the downscaled model data using CRU-observed gridded data for the historical period 1986-2005.The results indicate that the downscaled model data can generally reproduce the pattern characteristics of temperature and precipitation variations well over the major BRI regions for 1986-2005.Furthermore,we project temperature and precipitation variations over the major BRI regions at global temperature rise targets of 1.5°C,2°C,and 3°C under the RCP8.5 emission scenario based on the dataset by adopting the multiple-model ensemble mean.Our dataset contributes to understanding detailed the characteristics of climate change over the major BRI regions,and provides data fundamental for adopting appropriate strategies and options to reduce or avoid disadvantaged consequences associated with climate change over the major BRI regions.The dataset is available at https://doi.org/10.57760/sciencedb.01850.
