Search Results
You are looking at 1 - 10 of 31 items for
- Author or Editor: Jiang Zhu x
- Refine by Access: All Content x
Abstract
It is found that some stable time-integration schemes for some nonlinear models do not guarantee stable integrations of the associated tangent linear models with the same time step size. These problems usually occur when the nonlinear models describe vertical diffusion processes and are numerically implemented by semi-implicit time-integration schemes that are unconditionally stable. The direct linearization procedure performed on such numerical schemes of nonlinear models can be interpreted as some conditionally stable numerical schemes of the underlying linearized equations.
Numerical experiments using a simple, illustrative model and a realistic ocean mixed layer model and their tangent linear models showed instabilities in the tangent linear models. Several methods are tried to reduce the nonphysical noise caused by the numerical instabilities. This study suggests that reducing time step size can give good results compared to some other methods that either are not accurate enough or change too much from the original nonlinear model.
Abstract
It is found that some stable time-integration schemes for some nonlinear models do not guarantee stable integrations of the associated tangent linear models with the same time step size. These problems usually occur when the nonlinear models describe vertical diffusion processes and are numerically implemented by semi-implicit time-integration schemes that are unconditionally stable. The direct linearization procedure performed on such numerical schemes of nonlinear models can be interpreted as some conditionally stable numerical schemes of the underlying linearized equations.
Numerical experiments using a simple, illustrative model and a realistic ocean mixed layer model and their tangent linear models showed instabilities in the tangent linear models. Several methods are tried to reduce the nonphysical noise caused by the numerical instabilities. This study suggests that reducing time step size can give good results compared to some other methods that either are not accurate enough or change too much from the original nonlinear model.
Abstract
A complete map of the ocean subsurface temperature is essential for monitoring aspects of climate change such as the ocean heat content (OHC) and sea level changes and for understanding the dynamics of the ocean/climate variation. However, global observations have not been available in the past, so a mapping strategy is required to fill the data gaps. In this study, an advanced mapping method is proposed to reconstruct the historical ocean subsurface (0–700 m) temperature field from 1940 to 2014 by using ensemble optimal interpolation with a dynamic ensemble (EnOI-DE) approach and a multimodel ensemble of phase 5 of the Coupled Model Intercomparison Project (CMIP5) historical and representative concentration pathway 4.5 simulations. The reconstructed field is a combination of two parts: a first guess provided by the ensemble mean of CMIP5 models and an adjustment by minimizing the analysis error with the assistance of error covariance determined by the CMIP5 models. The uncertainty of the field can also be assessed. This new approach was evaluated using a series of tests, including subsample tests by using data from the Argo period, idealized tests by specifying a truth field from the models, and withdrawn-data tests by removing 20% of the observations for validation. In addition, the authors showed that the ocean mean state, long-term trends, and interannual and decadal variability are all well represented. Furthermore, the most significant benefit of this method is to provide an improved estimate of the long-term historical OHC changes since 1940, which have important implications for Earth’s energy budget.
Abstract
A complete map of the ocean subsurface temperature is essential for monitoring aspects of climate change such as the ocean heat content (OHC) and sea level changes and for understanding the dynamics of the ocean/climate variation. However, global observations have not been available in the past, so a mapping strategy is required to fill the data gaps. In this study, an advanced mapping method is proposed to reconstruct the historical ocean subsurface (0–700 m) temperature field from 1940 to 2014 by using ensemble optimal interpolation with a dynamic ensemble (EnOI-DE) approach and a multimodel ensemble of phase 5 of the Coupled Model Intercomparison Project (CMIP5) historical and representative concentration pathway 4.5 simulations. The reconstructed field is a combination of two parts: a first guess provided by the ensemble mean of CMIP5 models and an adjustment by minimizing the analysis error with the assistance of error covariance determined by the CMIP5 models. The uncertainty of the field can also be assessed. This new approach was evaluated using a series of tests, including subsample tests by using data from the Argo period, idealized tests by specifying a truth field from the models, and withdrawn-data tests by removing 20% of the observations for validation. In addition, the authors showed that the ocean mean state, long-term trends, and interannual and decadal variability are all well represented. Furthermore, the most significant benefit of this method is to provide an improved estimate of the long-term historical OHC changes since 1940, which have important implications for Earth’s energy budget.
Abstract
Sea level anomalies (SLA) from the Ocean Topography Experiment (TOPEX)/Poseidon are assimilated with three-dimensional variational data assimilation (3DVAR) and ensemble optimal interpolation (EnOI) for the period of 1997–2001. When sea level data are assimilated, one major concern is how to project the surface information downward. In 3DVAR, downward projection is usually achieved by minimizing a cost function that computes the relations among temperature, salinity, and sea level. In EnOI, the surface information is propagated to other variables through a stationary ensemble. Their effects on the simulated variability are evaluated in a tropical Pacific Ocean model. When compared with different datasets, it is found that effects of 3DVAR and EnOI are different in several aspects. For sea level, the standard deviation is improved by both methods, but EnOI is more effective in the central/eastern Pacific. The SLA evolution is better reproduced with EnOI than with 3DVAR. For temperature, the model–reanalysis correlations are increased by 0.1–0.2 in the top 200 m with both methods, but EnOI is more effective, especially along the thermocline depth. When compared with the Tropical Atmosphere–Ocean array (TAO) profiles, evolution of the temperature reveals that 3DVAR tends to cause more errors during ENSO events. The correlations with TAO profile are increased by 0.1–0.3 with EnOI and are generally decreased by 0.1–0.3 with 3DVAR. For salinity, both methods have weak impact on the model–reanalysis correlations above the thermocline. Relative to 3DVAR, EnOI can increase the correlation by 0.2 below the thermocline. When compared with the TAO profiles, the differences are reduced to some extent with both methods, but 3DVAR is very negative on the simulated variability.
Abstract
Sea level anomalies (SLA) from the Ocean Topography Experiment (TOPEX)/Poseidon are assimilated with three-dimensional variational data assimilation (3DVAR) and ensemble optimal interpolation (EnOI) for the period of 1997–2001. When sea level data are assimilated, one major concern is how to project the surface information downward. In 3DVAR, downward projection is usually achieved by minimizing a cost function that computes the relations among temperature, salinity, and sea level. In EnOI, the surface information is propagated to other variables through a stationary ensemble. Their effects on the simulated variability are evaluated in a tropical Pacific Ocean model. When compared with different datasets, it is found that effects of 3DVAR and EnOI are different in several aspects. For sea level, the standard deviation is improved by both methods, but EnOI is more effective in the central/eastern Pacific. The SLA evolution is better reproduced with EnOI than with 3DVAR. For temperature, the model–reanalysis correlations are increased by 0.1–0.2 in the top 200 m with both methods, but EnOI is more effective, especially along the thermocline depth. When compared with the Tropical Atmosphere–Ocean array (TAO) profiles, evolution of the temperature reveals that 3DVAR tends to cause more errors during ENSO events. The correlations with TAO profile are increased by 0.1–0.3 with EnOI and are generally decreased by 0.1–0.3 with 3DVAR. For salinity, both methods have weak impact on the model–reanalysis correlations above the thermocline. Relative to 3DVAR, EnOI can increase the correlation by 0.2 below the thermocline. When compared with the TAO profiles, the differences are reduced to some extent with both methods, but 3DVAR is very negative on the simulated variability.
Abstract
Assessment of the upper-ocean (0–700 m) heat content (OHC) is a key task for monitoring climate change. However, irregular spatial and temporal distribution of historical subsurface observations has induced uncertainties in OHC estimation. In this study, a new source of uncertainties in calculating OHC due to the insufficiency of vertical resolution in historical ocean subsurface temperature profile observations was diagnosed. This error was examined by sampling a high-vertical-resolution climatological ocean according to the depth intervals of in situ subsurface observations, and then the error was defined as the difference between the OHC calculated by subsampled profiles and the OHC of the climatological ocean. The obtained resolution-induced error appeared to be cold in the upper 100 m (with a peak of approximately −0.1°C), warm within 100–700 m (with a peak of ~0.1°C near 180 m), and warm when averaged over 0–700-m depths (with a global average of ~0.01°–0.025°C, ~1–2.5 × 1022 J). Geographically, it showed a warm bias within 30°S–30°N and a cold bias at higher latitudes in both hemispheres, the sign of which depended on the concave or convex shape of the vertical temperature profiles. Finally, the authors recommend maintaining an unbiased observation system in the future: a minimal vertical depth bin of 5% of the depth was needed to reduce the vertical-resolution-induced bias to less than 0.005°C on global average (equal to Argo accuracy).
Abstract
Assessment of the upper-ocean (0–700 m) heat content (OHC) is a key task for monitoring climate change. However, irregular spatial and temporal distribution of historical subsurface observations has induced uncertainties in OHC estimation. In this study, a new source of uncertainties in calculating OHC due to the insufficiency of vertical resolution in historical ocean subsurface temperature profile observations was diagnosed. This error was examined by sampling a high-vertical-resolution climatological ocean according to the depth intervals of in situ subsurface observations, and then the error was defined as the difference between the OHC calculated by subsampled profiles and the OHC of the climatological ocean. The obtained resolution-induced error appeared to be cold in the upper 100 m (with a peak of approximately −0.1°C), warm within 100–700 m (with a peak of ~0.1°C near 180 m), and warm when averaged over 0–700-m depths (with a global average of ~0.01°–0.025°C, ~1–2.5 × 1022 J). Geographically, it showed a warm bias within 30°S–30°N and a cold bias at higher latitudes in both hemispheres, the sign of which depended on the concave or convex shape of the vertical temperature profiles. Finally, the authors recommend maintaining an unbiased observation system in the future: a minimal vertical depth bin of 5% of the depth was needed to reduce the vertical-resolution-induced bias to less than 0.005°C on global average (equal to Argo accuracy).
Abstract
The choice of climatology is an essential step in calculating the key climate indicators, such as historical ocean heat content (OHC) change. The anomaly field is required during the calculation and is obtained by subtracting the climatology from the absolute field. The climatology represents the ocean spatial variability and seasonal circle. This study found a considerable weaker long-term trend when historical climatologies (constructed by using historical observations within a long time period, i.e., 45 yr) were used rather than Argo-period climatologies (i.e., constructed by using observations during the Argo period, i.e., since 2004). The change of the locations of the observations (horizontal sampling) during the past 50 yr is responsible for this divergence, because the ship-based system pre-2000 has insufficient sampling of the global ocean, for instance, in the Southern Hemisphere, whereas this area began to achieve full sampling in this century by the Argo system. The horizontal sampling change leads to the change of the reference time (and reference OHC) when the historical-period climatology is used, which weakens the long-term OHC trend. Therefore, Argo-period climatologies should be used to accurately assess the long-term trend of the climate indicators, such as OHC.
Abstract
The choice of climatology is an essential step in calculating the key climate indicators, such as historical ocean heat content (OHC) change. The anomaly field is required during the calculation and is obtained by subtracting the climatology from the absolute field. The climatology represents the ocean spatial variability and seasonal circle. This study found a considerable weaker long-term trend when historical climatologies (constructed by using historical observations within a long time period, i.e., 45 yr) were used rather than Argo-period climatologies (i.e., constructed by using observations during the Argo period, i.e., since 2004). The change of the locations of the observations (horizontal sampling) during the past 50 yr is responsible for this divergence, because the ship-based system pre-2000 has insufficient sampling of the global ocean, for instance, in the Southern Hemisphere, whereas this area began to achieve full sampling in this century by the Argo system. The horizontal sampling change leads to the change of the reference time (and reference OHC) when the historical-period climatology is used, which weakens the long-term OHC trend. Therefore, Argo-period climatologies should be used to accurately assess the long-term trend of the climate indicators, such as OHC.
Abstract
The ensemble Kalman filter (EnKF) has proven its efficiency in strongly nonlinear dynamical systems but is demanding in its computing power requirements, which are typically about the same as those of the four-dimensional variational data assimilation (4DVAR) systems presently used in several weather forecasting centers. A simplified version of EnKF, the so-called ensemble optimal interpolation (EnOI), requires only a small fraction of the computing cost of the EnKF, but makes the crude assumption of no dynamical evolution of the errors. How do both these two methods compare in realistic settings of a Pacific Ocean forecasting system where the computational cost is a primary concern? In this paper the two methods are used to assimilate real altimetry data via a Hybrid Coordinate Ocean Model of the Pacific. The results are validated against the independent Argo temperature and salinity profiles and show that the EnKF has the advantage in terms of both temperature and salinity and in all parts of the domain, although not with a very striking difference.
Abstract
The ensemble Kalman filter (EnKF) has proven its efficiency in strongly nonlinear dynamical systems but is demanding in its computing power requirements, which are typically about the same as those of the four-dimensional variational data assimilation (4DVAR) systems presently used in several weather forecasting centers. A simplified version of EnKF, the so-called ensemble optimal interpolation (EnOI), requires only a small fraction of the computing cost of the EnKF, but makes the crude assumption of no dynamical evolution of the errors. How do both these two methods compare in realistic settings of a Pacific Ocean forecasting system where the computational cost is a primary concern? In this paper the two methods are used to assimilate real altimetry data via a Hybrid Coordinate Ocean Model of the Pacific. The results are validated against the independent Argo temperature and salinity profiles and show that the EnKF has the advantage in terms of both temperature and salinity and in all parts of the domain, although not with a very striking difference.
Abstract
The use of high-density remote sensing buoys and ship-based observations play an increasingly crucial role in the operational assimilation and forecast of oceans. With the recent release of several high-resolution observation datasets, such as the Global Ocean Data Assimilation Experiment (GODAE) high-resolution SST (GHRSST) datasets, the development of observation-thinning schemes becomes important in the process of data assimilation because the huge quantity and dense spatial–temporal distributions of these datasets might make it expensive to assimilate the full dataset into ocean models or even decay the assimilation result. In this paper, an objective model simulation ensemble-based observation-thinning scheme is proposed and applied to a Chinese shelf–coastal seas eddy-resolving model. A successful thinning scheme should select a subset of observations yielding a small analysis error variance (AEV) while keeping the number of observations to as few as possible. In this study, the background error covariance (BEC) is estimated using the historical ensemble and then the subset of observations to minimize the AEV is selected, which is estimated from the Kalman theory. The authors used this method in the GHRSST product to cover the shelf and coastal seas around China and then verified the result with an estimation function and assimilation–forecast systems.
Abstract
The use of high-density remote sensing buoys and ship-based observations play an increasingly crucial role in the operational assimilation and forecast of oceans. With the recent release of several high-resolution observation datasets, such as the Global Ocean Data Assimilation Experiment (GODAE) high-resolution SST (GHRSST) datasets, the development of observation-thinning schemes becomes important in the process of data assimilation because the huge quantity and dense spatial–temporal distributions of these datasets might make it expensive to assimilate the full dataset into ocean models or even decay the assimilation result. In this paper, an objective model simulation ensemble-based observation-thinning scheme is proposed and applied to a Chinese shelf–coastal seas eddy-resolving model. A successful thinning scheme should select a subset of observations yielding a small analysis error variance (AEV) while keeping the number of observations to as few as possible. In this study, the background error covariance (BEC) is estimated using the historical ensemble and then the subset of observations to minimize the AEV is selected, which is estimated from the Kalman theory. The authors used this method in the GHRSST product to cover the shelf and coastal seas around China and then verified the result with an estimation function and assimilation–forecast systems.
Abstract
A new technique to estimate three major biases of XBT probes (improper fall rate, start-up transient, and pure temperature error) has been developed. Different from the well-known and standard “temperature error free” differential method, the new method analyses temperature profiles instead of vertical gradient temperature profiles. Consequently, it seems to be more noise resistant because it uses the integral property over the entire vertical profile instead of gradients. Its validity and robustness have been checked in two ways. In the first case, the new integral technique and the standard differential method have been applied to a set of simulated XBT profiles having a known fall-rate equation to which various combinations of pure temperature errors, random errors, and spikes have been added for the sake of this simulation. Results indicated that the single pure temperature error has little impact on the fall-rate coefficients for both methods, whereas with the added random error and spikes the simulation leads to better results with the new integral technique than with the standard differential method. In the second case, two sets of profiles from actual XBT versus CTD comparisons, collected near Barbados in 1990 and in the western Mediterranean (2003–04 and 2008–09), have been used. The individual fall-rate coefficients and start-up transient for each XBT profile, along with the overall pure temperature correction, have been calculated for the XBT profiles. To standardize procedures and to improve the terms of comparison, the individual start-up transient estimated by the integral method was also assigned and included in calculations with the differential method. The new integral method significantly reduces both the temperature difference between XBT and CTD profiles and the standard deviation. Finally, the validity of the mean fall-rate coefficients and the mean start-up transient, respectively, for DB and T7 probes as precalculated equations was verified. In this case, the temperature difference is reduced to less than 0.1°C for both datasets, and it randomly distributes around the null value. In addition, the standard deviation on depth values is largely reduced, and the maximum depth error computed with the datasets near Barbados is within 1.1% of its real value. Results also indicate that the integral method has a good performance mainly when applied to profiles in regions with either a very large temperature gradient, at the thermocline or a very small one, toward the bottom.
Abstract
A new technique to estimate three major biases of XBT probes (improper fall rate, start-up transient, and pure temperature error) has been developed. Different from the well-known and standard “temperature error free” differential method, the new method analyses temperature profiles instead of vertical gradient temperature profiles. Consequently, it seems to be more noise resistant because it uses the integral property over the entire vertical profile instead of gradients. Its validity and robustness have been checked in two ways. In the first case, the new integral technique and the standard differential method have been applied to a set of simulated XBT profiles having a known fall-rate equation to which various combinations of pure temperature errors, random errors, and spikes have been added for the sake of this simulation. Results indicated that the single pure temperature error has little impact on the fall-rate coefficients for both methods, whereas with the added random error and spikes the simulation leads to better results with the new integral technique than with the standard differential method. In the second case, two sets of profiles from actual XBT versus CTD comparisons, collected near Barbados in 1990 and in the western Mediterranean (2003–04 and 2008–09), have been used. The individual fall-rate coefficients and start-up transient for each XBT profile, along with the overall pure temperature correction, have been calculated for the XBT profiles. To standardize procedures and to improve the terms of comparison, the individual start-up transient estimated by the integral method was also assigned and included in calculations with the differential method. The new integral method significantly reduces both the temperature difference between XBT and CTD profiles and the standard deviation. Finally, the validity of the mean fall-rate coefficients and the mean start-up transient, respectively, for DB and T7 probes as precalculated equations was verified. In this case, the temperature difference is reduced to less than 0.1°C for both datasets, and it randomly distributes around the null value. In addition, the standard deviation on depth values is largely reduced, and the maximum depth error computed with the datasets near Barbados is within 1.1% of its real value. Results also indicate that the integral method has a good performance mainly when applied to profiles in regions with either a very large temperature gradient, at the thermocline or a very small one, toward the bottom.
Abstract
The Fengyun-3 series of satellites (FY-3) began in May 2008 with the launch of FY-3A. The onboard Microwave Humidity Sounders (MWHSs) provide vertical information about water vapor, which is important for numerical weather prediction (NWP). The noise equivalent delta temperature (NEDT) of the MWHS is higher than that of the Microwave Humidity Sounder (MHS) instrument (e.g., on board MetOp-B) but lower than that of the older AMSU-B instruments (on board NOAA-15, NOAA-16, and NOAA-17). Assimilation of MWHS observations into the ECMWF Integrated Forecast System (IFS) improved the fit of short-range forecasts to other observations, notably MHS, and also slightly improved the longer-range forecast scores verified against analyses. Also, assimilating the MWHS on board both FY-3A and FY-3B gave a larger impact than either instrument alone. Furthermore, when MWHS and MHS were added separately to a baseline using neither, the impact of MWHS was found to be comparable to that of MHS. Consequently, ECMWF has been assimilating the FY-3B MWHS data in the operational forecasting system since 24 September 2014. This is the first operational use of Chinese polar-orbiting satellite data by an NWP center outside of China.
Abstract
The Fengyun-3 series of satellites (FY-3) began in May 2008 with the launch of FY-3A. The onboard Microwave Humidity Sounders (MWHSs) provide vertical information about water vapor, which is important for numerical weather prediction (NWP). The noise equivalent delta temperature (NEDT) of the MWHS is higher than that of the Microwave Humidity Sounder (MHS) instrument (e.g., on board MetOp-B) but lower than that of the older AMSU-B instruments (on board NOAA-15, NOAA-16, and NOAA-17). Assimilation of MWHS observations into the ECMWF Integrated Forecast System (IFS) improved the fit of short-range forecasts to other observations, notably MHS, and also slightly improved the longer-range forecast scores verified against analyses. Also, assimilating the MWHS on board both FY-3A and FY-3B gave a larger impact than either instrument alone. Furthermore, when MWHS and MHS were added separately to a baseline using neither, the impact of MWHS was found to be comparable to that of MHS. Consequently, ECMWF has been assimilating the FY-3B MWHS data in the operational forecasting system since 24 September 2014. This is the first operational use of Chinese polar-orbiting satellite data by an NWP center outside of China.
Abstract
A seasonal evolution of rainbands over East China is evident and shows remarkable year-to-year variations. The present study identifies two dominant interannual modes of the seasonal evolution of rainbands over East China from 1981 to 2018: 1) the sudden change pattern, in which the anomalous rainfall changes abruptly from boreal spring to summer, especially over South China; and 2) the northward migration pattern, which shows a gradual poleward migration of the anomalous rainband over East China with the East Asian summer monsoon (EASM). Both of them are regulated by the sea surface temperature anomalies (SSTAs) in the Northern Hemisphere from spring to summer. In the sudden change pattern, the SSTAs in the Pacific modulate spring rainfall over South China via the ENSO–EASM teleconnection. By contrast, the North Atlantic SSTAs change the midlatitude wave train and modify summer rainfall over South and North China, in conjunction with the anomalous tropical circulation due to the Indian Ocean SSTAs. In the northward migration pattern, the North Pacific SSTAs alter spring rainfall over South China by varying the low-level western North Pacific subtropical high and the zonal land–sea thermal contrast over East Asia. Afterward, the ENSO-like SSTAs induce a Pacific–Japan teleconnection and shift the anomalous rainband northward to the Yangtze–Huai River and North China in summer. The seasonal switch of the SSTAs regulating these two modes is physically linked from boreal spring to summer. This mechanism provides potential seasonal predictability of the seasonal evolution of the anomalous rainband over East China.
Abstract
A seasonal evolution of rainbands over East China is evident and shows remarkable year-to-year variations. The present study identifies two dominant interannual modes of the seasonal evolution of rainbands over East China from 1981 to 2018: 1) the sudden change pattern, in which the anomalous rainfall changes abruptly from boreal spring to summer, especially over South China; and 2) the northward migration pattern, which shows a gradual poleward migration of the anomalous rainband over East China with the East Asian summer monsoon (EASM). Both of them are regulated by the sea surface temperature anomalies (SSTAs) in the Northern Hemisphere from spring to summer. In the sudden change pattern, the SSTAs in the Pacific modulate spring rainfall over South China via the ENSO–EASM teleconnection. By contrast, the North Atlantic SSTAs change the midlatitude wave train and modify summer rainfall over South and North China, in conjunction with the anomalous tropical circulation due to the Indian Ocean SSTAs. In the northward migration pattern, the North Pacific SSTAs alter spring rainfall over South China by varying the low-level western North Pacific subtropical high and the zonal land–sea thermal contrast over East Asia. Afterward, the ENSO-like SSTAs induce a Pacific–Japan teleconnection and shift the anomalous rainband northward to the Yangtze–Huai River and North China in summer. The seasonal switch of the SSTAs regulating these two modes is physically linked from boreal spring to summer. This mechanism provides potential seasonal predictability of the seasonal evolution of the anomalous rainband over East China.
