Abstract

Radiosonde temperature records contain valuable information for climate change research from the 1940s onward. Since they are affected by numerous artificial shifts, time series homogenization efforts are required. This paper introduces a new technique that uses time series of temperature differences between the original radiosonde observations (obs) and background forecasts (bg) of an atmospheric climate data assimilation system for homogenization.

These obs − bg differences, the “innovations,” are a by-product of the data assimilation process. They have been saved during the 40-yr ECMWF Re-Analysis (ERA-40) and are now available for each assimilated radiosonde record back to 1958. It is demonstrated that inhomogeneities in the obs time series due to changes in instrumentation can be automatically detected and adjusted using daily time series of innovations at 0000 and 1200 UTC.

The innovations not only reveal problems of the radiosonde records but also of the data assimilation system. Although ERA-40 used a frozen data assimilation system, the time series of the bg contains some breaks as well, mainly due to changes in the satellite observing system. It has been necessary to adjust the global mean bg temperatures before the radiosonde homogenization.

After this step, homogeneity adjustments, which can be added to existing raw radiosonde observations, have been calculated for 1184 radiosonde records. The spatiotemporal consistency of the global radiosonde dataset is improved by these adjustments and spuriously large day–night differences are removed. After homogenization the climatologies of the time series from certain radiosonde types have been adjusted. This step reduces temporally constant biases, which are detrimental for reanalysis purposes. Therefore the adjustments applied should yield an improved radiosonde dataset that is suitable for climate analysis and particularly useful as input for future climate data assimilation efforts. The focus of this paper relies on the lower stratosphere and on the internal consistency of the homogenized radiosonde dataset. Implications for global mean upper-air temperature trends are touched upon only briefly.

Abstract

The vertically integrated global energy budget is evaluated with a direct and an indirect method (both corrected for mass inconsistencies of the forecast model), mainly using the European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis Interim (ERA-Interim) data. A new estimate for the net poleward total energy transport is given. Comparison to satellite-derived radiation data proves that ERA-Interim is better suited for investigation of interannual variations of the global energy budget than available satellite data since these either cover a relatively short period of time or are too inhomogeneous in time. While much improved compared to the 40-yr ECMWF Re-Analysis (ERA-40), regionally averaged energy budgets of ERA-Interim show that strong anomalies of forecasted vertical fluxes tend to be partly compensated by unrealistically large forecasted energy storage rates. Discrepancies between observed and forecasted monthly mean tendencies can be taken as rough measure for the uncertainties involved in the ERA-Interim energy budget. El Niño–Southern Oscillation (ENSO) is shown to have large impact on regional energy budgets, but strong compensation occurs between the western and eastern Pacific, leading to only small net variations of the total poleward energy transports (similar magnitude as the uncertainty of the computations). However, Hovmöller longitude–time plots of tropical energy exports show relatively strong slowly eastward-moving poleward transport anomalies in connection with ENSO. Verification of these findings using independent estimates still needs to be done.

Abstract

This article describes progress in the homogenization of global radiosonde temperatures with updated versions of the Radiosonde Observation Correction Using Reanalyses (RAOBCORE) and Radiosonde Innovation Composite Homogenization (RICH) software packages. These are automated methods to homogenize the global radiosonde temperature dataset back to 1958. The break dates are determined from analysis of time series of differences between radiosonde temperatures (obs) and background forecasts (bg) of climate data assimilation systems used for the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40) and the ongoing interim ECMWF Re-Analysis (ERA-Interim).

RAOBCORE uses the obs−bg time series also for estimating the break sizes. RICH determines the break sizes either by comparing the observations of a tested time series with observations of neighboring radiosonde time series (RICH-obs) or by comparing their background departures (RICH-τ). Consequently RAOBCORE results may be influenced by inhomogeneities in the bg, whereas break size estimation with RICH-obs is independent of the bg. The adjustment quality of RICH-obs, on the other hand, may suffer from large interpolation errors at remote stations. RICH-τ is a compromise that substantially reduces interpolation errors at the cost of slight dependence on the bg.

Adjustment uncertainty is estimated by comparing the three methods and also by varying parameters in RICH. The adjusted radiosonde time series are compared with recent temperature datasets based on (Advanced) Microwave Sounding Unit [(A)MSU] radiances. The overall spatiotemporal consistency of the homogenized dataset has improved compared to earlier versions, particularly in the presatellite era. Vertical profiles of temperature trends are more consistent with satellite data as well.

Abstract

The apparent cooling trend in observed global mean temperature series from radiosonde records relative to Microwave Sounding Unit (MSU) radiances has been a long-standing problem in upper-air climatology. It is very likely caused by a warm bias of radiosonde temperatures in the 1980s, which has been reduced over time with better instrumentation and correction software. The warm bias in the MSU-equivalent lower stratospheric (LS) layer is estimated as 0.6 ± 0.3 K in the global mean and as 1.0 ± 0.3 K in the tropical (20°S–20°N) mean. These estimates are based on comparisons of unadjusted radiosonde data, not only with MSU data but also with background forecast (BG) temperature time series from the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40) and with two new homogenized radiosonde datasets. One of the radiosonde datasets [Radiosonde Observation Correction using Reanalyses (RAOBCORE) version 1.4] employs the BG as reference for homogenization, which is not strictly independent of MSU data. The second radiosonde dataset uses the dates of the breakpoints detected by RAOBCORE as metadata for homogenization. However, it relies only on homogeneous segments of neighboring radiosonde data for break-size estimation. Therefore, adjustments are independent of satellite data.

Both of the new adjusted radiosonde time series are in better agreement with satellite data than comparable published radiosonde datasets, not only for zonal means but also at most single stations. A robust warming maximum of 0.2–0.3K (10 yr)⁻¹ for the 1979–2006 period in the tropical upper troposphere could be found in both homogenized radiosonde datasets. The maximum is consistent with mean temperatures of a thick layer in the upper troposphere and upper stratosphere (TS), derived from M3U3 radiances. Inferred from these results is that it is possible to detect and remove most of the mean warm bias from the radiosonde records, and thus most of the trend discrepancy compared to MSU LS and TS temperature products.

The comprehensive intercomparison also suggests that the BG is temporally quite homogeneous after 1986. Only in the early 1980s could some inhomogeneities in the BG be detected and quantified.

Abstract

This study uses advanced numerical and diagnostic methods to evaluate the atmospheric energy budget with the fifth major global reanalysis produced by ECMWF (ERA5) in combination with observed and reconstructed top of the atmosphere (TOA) energy fluxes for the period 1985–2018. We assess the meridional as well as ocean–land energy transport and perform internal consistency checks using mass-balanced data. Furthermore, the moisture and mass budgets in ERA5 are examined and compared with previous budget evaluations using ERA-Interim as well as observation-based estimates. Results show that peak annual mean meridional atmospheric energy transports in ERA5 (4.58 ± 0.07 PW in the Northern Hemisphere) are weaker compared to ERA-Interim (4.74 ± 0.09 PW), where the higher spatial and temporal resolution of ERA5 can be excluded as a possible reason. The ocean–land energy transport in ERA5 is reliable at least from 2000 onward (~2.5 PW) such that the imbalance between net TOA fluxes and lateral energy fluxes over land are on the order of ~1 W m⁻². Spinup and spindown effects as revealed from inconsistencies between analyses and forecasts are generally smaller and temporally less variable in ERA5 compared to ERA-Interim. Evaluation of the moisture budget shows that the ocean–land moisture transport and parameterized freshwater fluxes agree well in ERA5, while there are large inconsistencies in ERA-Interim. Overall, the quality of the budgets derived from ERA5 is demonstrably better than estimates from ERA-Interim. Still some particularly sensitive budget quantities (e.g., precipitation, evaporation, and ocean–land energy transport) show apparent inhomogeneities, especially in the late 1990s, which warrant further investigation and need to be considered in studies of interannual variability and trends.

Abstract

Snow cover duration is commonly derived from snow depth, snow water equivalent, or satellite data. Snow cover duration has more recently also been inferred from ground temperature data. In this study, a probabilistic snow cover duration (SCD) model is introduced that estimates the conditional probability for snow cover given the daily mean and the diurnal range of ground temperature. For the application of the SCD model, 87 Austrian sites in the Alpine region are investigated in the period of 2000 to 2011. The daily range of ground temperature is identified to represent the primary variable in determining the snow cover duration. In the case of a large dataset, however, the inclusion of the daily mean ground temperature as the second given parameter improves results. Rank correlation coefficients of predicted versus observed snow cover duration are typically between 0.8 and 0.9.

Abstract

Vast amounts of energy are exchanged between the ocean, atmosphere, and space in association with El Niño–Southern Oscillation (ENSO). This study examines energy budgets of all tropical (30°S–30°N) ocean basins and the atmosphere separately using different, largely independent oceanic and atmospheric reanalyses to depict anomalous energy flows associated with ENSO in a consistent framework. It is found that variability of area-averaged ocean heat content (OHC) in the tropical Pacific to a large extent is modulated by energy flow through the ocean surface. While redistribution of OHC within the tropical Pacific is an integral part of ENSO dynamics, variability of ocean heat transport out of the tropical Pacific region is found to be mostly small. Noteworthy contributions arise from the Indonesian Throughflow (ITF), which is anticorrelated with ENSO at a few months lag, and from anomalous oceanic poleward heat export during the La Niña events in 1999 and 2008. Regression analysis reveals that atmospheric energy transport and radiation at the top of the atmosphere (Rad_TOA) almost perfectly balance the OHC changes and ITF variability associated with ENSO. Only a small fraction of El Niño–related heat lost by the Pacific Ocean through anomalous air–sea fluxes is radiated to space immediately, whereas the major part of the energy is transported away by the atmosphere. Ample changes in tropical atmospheric circulation lead to enhanced surface fluxes and, consequently, to an increase of OHC in the tropical Atlantic and Indian Ocean that almost fully compensates for tropical Pacific OHC loss. This signature of energy redistribution is robust across the employed datasets for all three tropical ocean basins and explains the small ENSO signal in global mean Rad_TOA.

Abstract

This study uses the ECMWF ERA5 reanalysis and observationally constrained top-of-the-atmosphere radiative fluxes to infer net surface energy fluxes covering 1985–2018, which can be further adjusted to match the observed mean land heat uptake. Various diagnostics are applied to provide error estimates of inferred fluxes on different spatial scales. For this purpose, adjusted as well as unadjusted inferred surface fluxes are compared with other commonly used flux products. On a regional scale, the oceanic energy budget of the North Atlantic between the RAPID array at 26.5°N and moorings located farther north (e.g., at the Greenland–Scotland Ridge) is evaluated. On the station scale, a comprehensive comparison of inferred and buoy-based fluxes is presented. Results indicate that global land and ocean averages of unadjusted inferred surface fluxes agree with the observed heat uptake to within 1 W m⁻², while satellite-derived and model-based fluxes show large global mean biases. Furthermore, the oceanic energy budget of the North Atlantic is closed to within 2.7 (−0.2) W m⁻² for the period 2005–09 when unadjusted (adjusted) inferred surface fluxes are employed. Indirect estimates of the 2004–16 mean oceanic heat transport at 26.5°N are 1.09 PW (1.17 PW with adjusted fluxes), which agrees well with observed RAPID transports. On the station scale, inferred fluxes exhibit a mean bias of −20.1 W m⁻² when using buoy-based fluxes as reference, which confirms expectations that biases increase from global to local scales. However, buoy-based fluxes as reference are debatable, and are likely positively biased, suggesting that the station-scale bias of inferred fluxes is more likely on the order of −10 W m⁻².

Abstract

The variability of zonally resolved tropical energy budgets in association with El Niño–Southern Oscillation (ENSO) is investigated. The most recent global atmospheric reanalyses from 1979 to 2011 are employed with removal of apparent discontinuities to obtain best possible temporal homogeneity. The growing length of record allows a more robust analysis of characteristic patterns of variability with cross-correlation, composite, and EOF methods. A quadrupole anomaly pattern is found in the vertically integrated energy divergence associated with ENSO, with centers over the Indian Ocean, the Indo-Pacific warm pool, the eastern equatorial Pacific, and the Atlantic. The smooth transition, particularly of the main maxima of latent and dry static energy divergence, from the western to the eastern Pacific is found to require at least two EOFs to be adequately described. The canonical El Niño pattern (EOF-1) and a transition pattern (EOF-2; referred to as El Niño Modoki by some authors) form remarkably coherent ENSO-related anomaly structures of the tropical energy budget not only over the Pacific but throughout the tropics. As latent and dry static energy divergences show strong mutual cancellation, variability of total energy divergence is smaller and more tightly coupled to local sea surface temperature (SST) anomalies and is mainly related to the ocean heat discharge and recharge during ENSO peak phases. The complexity of the structures throughout the tropics and their evolution during ENSO events along with their interactions with the annual cycle have often not been adequately accounted for; in particular, the El Niño Modoki mode is but part of the overall evolutionary patterns.