European Centre For Medium-Range Weather Forecasts Winds Research Articles

Eastern Mediterranean Transient studied with Lagrangian diagnostics applied to a Mediterranean OGCM forced by satellite SST and ECMWF wind stress for the years 1988–1993

In this work, we estimate deep and intermediate water mass dispersal in the different phases of the Eastern Mediterranean Transient by means of the Lagrangian analysis of a Mediterranean ocean general circulation model (OGCM) run, forced by daily European Centre for Medium‐Range Weather Forecasts (ECMWF) wind stress and satellite sea surface temperature (SST) over the years from 1988 to 1993. While in a first phase (1989–1991) major modifications of the intermediate salty waters path are observed, during 1992 and 1993 the general mechanism of the transient is reproduced in the deep Ionian basin, where the Cretan overflow replaces Adriatic deep water that uplifting of several hundred meters freshens in the intermediate layers. Analysis of the water mass formation rates shows that dense water production during 1993 is twice the climatological values in the eastern subbasins of the Mediterranean and that the densest water is formed in the Cretan Sea. From 1988 to 1993, about 1.3 × 1014 m3 of water flows over the Cretan Arcs (see the caption of Figure 1 for a definition of the Cretan Arcs) to sink deeper than 600 m in the eastern Mediterranean, which is roughly half of the value estimated by Roether et al. [1996]. About 60% of this deep outflow (∼0.8 × 1014 m3) occurs in a few months, during the cold winters of 1992 and 1993, when the overflow of Cretan dense water in the Antikithera Strait reaches 1.5 Sv and rapidly propagates in the deep Ionian basin, developing very energetic coherent structures. We study the characteristic times of the relaxation of the system toward the climatology, and finally, we estimate the rate and the geographical distribution of the upwelling in the intermediate and subsurface layers.

Implications of interannual variability in atmospheric circulation on modeled CO2 concentrations and source estimates

The impact of the interannual variability (IAV) of atmospheric transport on atmospheric CO2 observations is often ignored in carbon cycle studies. We use 8 years of analyzed European Centre for Medium‐Range Weather Forecasts (ECMWF) wind fields from 1985–1992 to demonstrate the effect of IAV of the circulation on modeled CO2 concentrations for fossil and biosphere CO2 sources. The wind fields are used to drive the Melbourne University Tracer Model. The modeled annual mean CO2 values at observing locations show little IAV in the fossil case. In the biosphere case the IAV of the annual mean is considerably larger, especially in the high northern latitudes. Simulated baseline selection using modeled radon concentrations does not reduce the IAV. In both cases, surface layer interhemispheric differences show small interannual variations indicating small changes in interhemispheric transport. Source fields from a mass balance inversion using 1985–1992 ECMWF winds are compared with those from inversions using a single year of winds repeated for each year of observations. We find differences in semi‐hemispheric sources of up to 0.6 Gt C yr−1 at some times. However, 8 year mean sources show smaller differences, mostly less than 0.2 Gt C yr−1 at regional scales. This indicates that the variability in the circulation is a second order factor in determining the sources from inversion calculations.

Evaluation of NSCAT-2 Wind Vectors by Using Statistical Distributions of Wind Speeds and Directions

In order to validate wind vectors derived from the NASA scatterometer (NSCAT), statistical distributions of wind speeds and directions retrieved by the NSCAT-2 geophysical model function have been investigated by comparison with wind data retrieved by the other model functions such as SASS-2 and NSCAT-1 and those derived from the wind analyses of the European Centre for Medium Range Weather Forecasts (ECMWF). The histogram of the NSCAT-2 wind speeds has a similar shape to those of the ECMWF and NSCAT-1 winds, but is slightly shifted toward higher wind speed to adjust negative bias which has been found in the NSCAT-1 winds by previous buoy comparison studies. Variations of the standard deviation of the NSCAT-2 wind speeds with incidence angle are greater than those of the ECMWF and NSCAT-1 winds. The frequency distribution of wind directions relative to spacecraft flight direction has been calculated to assess the self-consistency of the wind directions. It was found that the NSCAT-2 wind vectors exhibit systematic directional preference relative to antenna beams. This artificial directivity is considered to be caused by imperfections in the antenna beam balancing and the geophysical model function. The skill of the ambiguity removal procedure is discussed as a function of wind speed and incidence angle, and is found to be improved compared to the NSCAT-1 winds, especially at high incidence angles. It is concluded that systematic errors in wind directions might be increased by modifications from NSCAT-1 to NSCAT-2, though the wind speed bias is removed and the ambiguity removal skill is improved.

A model function for the ocean‐normalized radar cross section at 14 GHz derived from NSCAT observations

A model for the ocean surface normalized radar cross section σo is derived from 3 months of NASA scatterometer (NSCAT) observations (September 15 to December 18, 1996). The model expresses σo as a function of wind speed, relative wind direction, incidence angle, and polarization. The dependence of σo on wind speed is based on collocated special sensor microwave/imager (SSM/I) satellite wind retrievals and European Centre for Medium‐Range Weather Forecasts (ECMWF) model winds. We find that at low winds (<5 ms−1), the SSM/I winds are more reliable than ECWMF, probably owing to small location errors in the ECMWF wind features. The primary wind direction dependence of σo (i.e., the second harmonic) is found from histograms of the σo difference between the forward and aft antennas. The σo versus wind speed relationship is adjusted for cross‐swath incidence angle differences and is then incorporated into the NSCAT 1 model used to process the 10‐month (September 15, 1996, to June 29, 1997) NSCAT data set. The resulting NSCAT 1 wind vectors are compared to ECMWF wind fields and buoys. The mean and standard deviation of the NSCAT minus ECMWF (buoy) wind speed difference are 0.05 and 1.78 ms−1 (−0.29 and 1.26 ms−1), respectively. The wind direction mean and standard deviation differences are 0.8° and 18.5° (7.9° and 15.7°), respectively. The difference between the NSCAT and the ECMWF (buoy) direction exceeds 90° only 1.1% (1.2%) of the time. We have no explanation for why the buoy wind directions are biased 8° relative to both NSCAT and ECMWF.

Statistical distribution of wind speeds and directions globally observed by NSCAT

In order to validate wind vectors derived from the NASA scatterometer (NSCAT), statistical distributions of wind speeds and directions over the global oceans are investigated by comparing with European Centre for Medium‐Range Weather Forecasts (ECMWF) wind data. Histograms of wind speeds and directions are calculated from the preliminary and reprocessed NSCAT data products for a period of 8 weeks. For wind speed of the preliminary data products, excessive low wind distribution is pointed out through comparison with ECMWF winds. A hump at the lower wind speed side of the peak in the wind speed histogram is discernible. The shape of the hump varies with incidence angle. Incompleteness of the prelaunch geophysical model function, SASS 2, tentatively used to retrieve wind vectors of the preliminary data products, is considered to cause the skew of the wind speed distribution. On the contrary, histograms of wind speeds of the reprocessed data products show consistent features over the whole range of incidence angles. Frequency distribution of wind directions relative to spacecraft flight direction is calculated to assess self‐consistency of the wind directions. It is found that wind vectors of the preliminary data products exhibit systematic directional preference relative to antenna beams. This artificial directivity is also considered to be caused by imperfections in the geophysical model function. The directional distributions of the reprocessed wind vectors show less directivity and consistent features, except for very low wind cases.

NSCAT tropical wind stress maps: Implications for improving ocean modeling

Using wind vectors from the NASA scatterometer (NSCAT), daily maps of pseudostress have been constructed for the tropical Pacific Ocean and compared with pseudostress maps derived from the European Centre for Medium‐Range Weather Forecasts (ECMWF) 10‐m wind product. The map resolution for the NSCAT pseudostress maps was selected using both a statistical measure of the expected mapping errors and tests on realistic wind fields. The selected map resolution is 5 days and 2°, which minimizes residual effects from the NSCAT sampling pattern, while maximizing temporal and spatial resolution. Comparisons with the ECMWF maps showed significant differences in most regions, corresponding to mean wind speeds of 2–3 m s−1, particularly in the Intertropical Convergence Zone (ITCZ) and at 20°S and 20°N. A canonical correlation analysis between NSCAT and ECMWF fields showed a high degree of correlation of temporal variations and systematic differences in spatial structure. In the NSCAT fields the ITCZ is narrower, stronger, and is located 1–2° latitude farther south than in the ECMWF fields. The high degree of correlation between the two fields suggests that “hybrid” fields can be generated. The dynamical implications of the differences in wind forcing are illustrated using estimates of the Sverdrup stream function and the Ekman pumping. A simple reduced‐gravity, linear vorticity model, forced by both the ECMWF and hybrid winds to examine predicted differences in ocean response, showed higher skill for the NSCAT winds.

Ocean Surface Winds off the West Coast of New Zealand: A Comparison of Ocean Buoy, ECMWF Model, and Land-Based Data

Abstract Wind data for the period July–October 1994 from an ocean buoy moored off the west coast of New Zealand have been compared with model wind data from the European Centre for Medium-Range Weather Forecasts (ECMWF) and data from the nearest land-based station. The buoy and ECMWF winds were highly coherent and strongly polarized in the alongshore direction, suggesting that the ECMWF model correctly accounts for the orographic effect of the mountainous New Zealand landmass on the prevailing synoptic weather systems. At low frequencies, the ECMWF data can be used as a substitute for in situ wind data in oceanographic applications in the New Zealand region. This is a similar result to that reported in some Northern Hemisphere studies. In contrast, winds at the land-based site were much weaker and less strongly aligned alongshore than winds recorded offshore.

Transport of 222radon to the remote troposphere using the Model of Atmospheric Transport and Chemistry and assimilated winds from ECMWF and the National Center for Environmental Prediction/NCAR

The Model of Atmospheric Transport and Chemistry (MATCH) is used to simulate the transport of 222Rn using both European Centre for Medium‐Range Weather Forecasts (ECMWF) winds and National Center for Environmental Prediction/National Center for Atmospheric Research (hereafter referred to as NCEP) reanalysis winds. These winds have the advantage of being based on observed winds but have the disadvantage that the subgrid‐scale transport processes are not routinely archived. MATCH derives subgrid‐scale mixing rates for the boundary layer using a nonlocal scheme and for moist convective mixing using one of two parameterizations (Tiedtke [1989] or Pan and Wu [1997]). This paper describes the ability of the model to recreate mixing rates of 222Rn using the forecast center winds. Radon 222 is a species with a continental crust source and a simple sink involving radioactive decay with an e‐folding timescale of 5.5 days. This atmospheric constituent is therefore a good tracer for testing the vertical transport in the chemical transport model, as well as the horizontal transport from continental regions to remote oceanic regions. The various simulations of 222Rn are compared with observations as well as with each other, allowing an estimate of the uncertainty in transport due to uncertainties in the winds and subgrid‐scale processes. The calculated vertical profiles over the western United States are somewhat similar to observed, and the upper tropospheric concentrations compare reasonably well in their spatial distribution with data collected during Tropospheric Ozone II (TROPOZ II), although the model values tend to be higher than observed values, especially in the upper troposphere. The model successfully simulates specific observed pollution events at Cape Grim. It has more difficulty at sites farther from continental source regions, although the model captures the seasonal structure of the pollution events at these sites (Macquarie Island, Amsterdam Island, Kerguelen Island, and Crozet Island). Inclusion of a moist convective mixing scheme in MATCH increases 222Rn concentrations in the upper troposphere by 50% compared to not having moist convective mixing, while surface concentrations do not appear to be very sensitive to moist convection. In addition, differences between the upper tropospheric concentrations of radon predicted using the ECMWF and NCEP winds can be 30% for large areas of the globe, due to either differences in the forecast center winds themselves or the moist convective mixing schemes used in conjunction with them. This has implications for model simulations of radiatively and chemically important trace species in the atmosphere.

Verification of the ECMWF Wave Forecasting System against Buoy and Altimeter Data

The present status of ocean wave modeling at the European Centre for Medium-Range Weather Forecasts (ECMWF) is reviewed. Ocean waves are forecasted globally up to 10 days by means of the Wave Model (WAM), which is driven by 10-m winds from the ECMWF atmospheric model. Initial conditions are provided by assimilation of ERS-1 data into the first-guess wave field. The analyzed wave height and peak period field are verified against buoy data and show a considerable improvement compared to verification results of a decade ago. This is confirmed by a comparison of first-guess wave height against ERS-1 altimeter data. The main reasons for this improvement are (i) the higher quality of ECMWF winds compared to a decade ago, (ii) the improved physics of the WAM model, and (iii) the assimilation of ERS-1 data. The forecast skill of the ECMWF wave forecasting system is also studied by comparing forecasts with buoy data and verifying analysis. Error growth in forecast wave height is less rapid than in forecast wind speed. However, considerable positive mean errors in forecast wave height are found, suggesting a too active atmospheric model in later stages of the forecast. Nevertheless, judging from anomaly correlation scores, the wave forecast seems to be useful up to day 5 in the forecast in the Northern Hemisphere. Since the wave forecast depends in a sensitive manner on the wind forecast, this confirms the high quality of ECMWF forecasts near the surface. Finally, promising ways of improving the wave forecast are also discussed, and, as an example, the positive impact three-dimensional variational assimilation in the atmospheric model has on the wave product is also mentioned.

Impact of Scatterometer Winds on Hydrologic Forcing and Convective Heating through Surface Divergence

Abstract One of the difficulties in estimating atmospheric heat and moisture budgets lies in resolving the near-surface heat and moisture convergence and vertical velocities, each of which is highly dependent on the divergence of the surface wind. A kinematic approach is proposed to utilize scatterometer winds to improve the estimate of surface wind divergence, the omega profile, and, therefore, estimates of the hydrologic forcing and convective heating over the tropical ocean. Improvements in these estimates over those obtained using analyses of the European Centre for Medium-Range Weather Forecasts (ECMWF) alone are found when the ERS-1 scatterometer ground tracks passed over the TOGA COARE intensive flux array. The spatial patterns of precipitation estimated from GMS IR temperatures and radar reflectivities agree better with the divergence fields derived from the ERS-1 scatterometer winds than with those derived from either ECMWF or rawinsonde winds. The use of ERS-1 surface winds also changes the omeg...

Comparison of ECMWF and satellite ocean wind speeds from 1985 to 1992

Abstract We compare wind speeds derived from analyses of the European Centre for Medium-Range Weather Forecasts (ECMWF) from 1986 to 1992 with wind speeds retrieved from three space borne instruments: the Geosat altimeter, the Special Sensor Microwave/Imager (SSM/I) microwave radiometer and the European Remote Sensing (ERS-I) scatterometer. The comparison reveals imperfections in both ECMWF and SSM/I wind speeds. The ECMWF wind speeds are systematically underestimated with respect to the satellite wind speeds in the tropical Pacific and the Atlantic Ocean before May 1989. After the changes to the ECMWF model in May 1989, the differences vanish in the Atlantic and are reduced by about 50 per cent in the Pacific Ocean. However, the differences between ECMWF and satellite wind speeds are greater than those observed between the satellite wind speeds themselves: yearly differences between the satellite and the ECMWF wind speeds are greater than 2 m s−1 in tropical regions whereas intercomparison of satellite w...

Responses of the tropical Pacific to wind forcing as observed by spaceborne sensors and simulated by an ocean general circulation model

In this study, satellite observations, in situ measurements, and model simulations are combined to assess the oceanic response to surface wind forcing in the equatorial Pacific. The surface wind fields derived from observations by the spaceborne special sensor microwave imager (SSM/I) and from the operational products of the European Centre for Medium‐Range Weather Forecasts (ECMWF) are compared. When SSM/I winds are used to force a primitive‐equation ocean general circulation model (OGCM), they produce 3°C more surface cooling than ECMWF winds for the eastern equatorial Pacific during the cool phase of an El Niño‐Southern Oscillation event. The stronger cooling by SSM/I winds is in good agreement with measurements at the moored buoys and observations by the advanced very high resolution radiometer, indicating that SSM/I winds are superior to ECMWF winds in forcing the tropical ocean. In comparison with measurements from buoys, tide gauges, and the Geosat altimeter, the OGCM simulates the temporal variations of temperature, steric, and sea level changes with reasonable realism when forced with the satellite winds. There are discrepancies between model simulations and observations that are common to both wind forcing fields, one of which is the simulation of zonal currents; they could be attributed to model deficiencies. By examining model simulations under two winds, vertical heat advection and uplifting of the thermocline are found to be the dominant factors in the anomalous cooling of the ocean mixed layer.

Three dimensional atmospheric transport simulation of the radioactive tracers 210Pb, 7Be, 10Be, and 90Sr

A global three‐dimensional transport model of the atmosphere, having a grid resolution of 7.83° (latitude) × 10° (longitude) × 19 layers in the vertical and driven by European Centre for Medium Range Weather Forecasts (ECMWF) wind field analyses of 1990, is used to simulate the distribution of natural (210Pb, 7Be, 10Be) and artificial (90Sr), radioactive, water‐soluble, aerosol‐borne tracers. Because of their well‐known source‐sink distribution these tracers build an ideal tool to depict transport processes in the whole atmosphere and to test the models ability of reproducing these. In particular, this paper focuses on mass exchange between stratosphere and troposphere, using the concentration ratio 10Be/7Be as an indicator. In general, the agreement between observations and model results is quite good, except in polar regions. Modeled arctic 210Pb and 7Be concentrations are overestimated and the annual cycle is underestimated. The modeled antarctic annual cycle of 10Be/7Be shows maxima in winter and spring, whereas the observations exhibit a summer maximum. These discrepancies are attributed to deficiencies in ECMWF wind fields, to differences between observed and model used precipitation, and to the employed parameterization schemes of dry and wet deposition which may be inadequate in the cold polar regions.

A Numerical Simulation of the Variability in the Tropical Atlantic Ocean, 1980–88

Abstract The authors have examined a numerical simulation of the tropical Atlantic Ocean circulation forced by the European Centre for Medium-Range Weather Forecasts (ECMWF) surface wind stress during 1980–88. The mean state and annual cycle of the ocean are realistically simulated by the model. The simulated interannual variability of sea surface temperature (SST) is also remarkably consistent with the observations, particularly the observed patterns of the basinwide warm/cold periods and the variation of dipole pattern and the associated meridional SST gradient. Discrepancies between observed and simulated SST anomalies are large early in the simulation, which seems caused by errors in the ECMWF wind analysis during that period. The low-frequency fluctuations of the meridional SST gradient associated with the dipole pattern during 1980–88 were caused by opposite SST anomalies between hemispheres, forced by out of phase fluctuations of the trade winds. Specifically, the southeast trade winds were anomalo...

The use of satellite scatterometer winds to drive a primitive equation model of the Indian Ocean: The impact of bandlike sampling

The aim of this study is to evaluate the impact of the bandlike sampling of spaceborne scatterometers on the ability of scatterometer winds to successfully force the mean flow and seasonal cycle of an ocean model in the context of equatorial and tropical dynamics. The equatorial ocean is simulated with a four‐layer, primitive equation, reduced gravity model of the Indian Ocean. The variable wind stress used in this study is derived from one year (1988) of 6‐hour analyses of the 10‐m wind vector over the Indian Ocean performed at the European Centre for Medium‐Range Weather Forecasts (ECMWF). It is applied as a forcing at every grid point of the model to drive a reference circulation. Scatterometer winds are simulated from ECMWF winds, using the nominal configurations and orbital parameters of the European Remote Sensing 1 (ERS‐1) and NASA Scatterometer (NSCAT) missions. The model is forced in real time under swaths with the raw scatterometer winds of ERS‐1 and NSCAT, with a persistence condition (i.e., the wind is kept constant until the next passage of the satellite provides a new value). The circulation obtained for each of the scatterometer experiments is compared with the reference circulation. The seasonal circulation of the Indian Ocean with NSCAT winds is very similar to the reference. The perturbations introduced by the bandlike sampling and the persistence condition have an impact similar to that of a small uncorrelated noise added to the reference forcing. The persistence condition for ERS‐1 does not give results which are as good as those obtained for NSCAT.

An Evaluation of ECMWF-Based Climatological Wind Stress Fields

A new mean monthly wind stress climatology based on seven yeas (1980–1986) of operational weather analyses by the European Centre for Medium-Range Weather Forecasts (ECMWF) has been derived by Trenberth et al. This climatology (referred to here as the TLO climatology) potentially represents a significant improvement over climatologies derived only from conventional wind observations. An attempt is made here to quantify the absolute accuracy of the TLO climatology by comparison with global wind stress fields constructed from vector winds measured by the Seasat-A Satellite Scatterometer (SASS) during 1978. From a simulated SASS dataset, it is shown that the magnitudes of the SASS stresses must be increased by about 7% to account for a systematic error that can be attributed to the scatterometer spatial- and temporal-sampling characteristics. After applying this correction, differences between the TLO climatology and SASS winds in the tropics are most likely related to known limitations of the ECMWF analyses. At latitudes south of 50°S, interannual variability and uncertainties in the operational weather analyses are so large that it is not possible to evaluate the TLO climatology on the basis of comparisons with SASS data. Outside of these equatorial and high southern latitude bands, the TLO stresses are shown to be systematically stronger than SASS by almost 50%. It is found that this difference can be entirely accounted for if the 1980–1986 ECMWF 1000-mb analyses are not interpreted as 10-m winds, as they were in constructing the TLO climatology. This conclusion is supported by an independent comparison of the synoptic ECMWF wind speed estimates with coincident buoy observations.

Determining torques over the ocean and their role in the planetary momentum budget

Knowledge of wind torques at the atmosphere‐ocean interface is important in understanding the global angular momentum balance. A surface wind data set that combines special sensor microwave/imager (SSM/I) and European Centre for Medium Range Weather Forecasts (ECMWF) wind products and covers the period July 1987 to June 1989 is used to estimate zonal torques over the global ocean. Comparison of these fields and fields computed from ECMWF wind analyses alone shows substantial differences in the statistics of both locally and globally integrated torques; the impact of SSM/I data on the torque analysis is largest over the Indian Ocean and eastern Pacific regions. The characteristics of the torques from the combined SSM/I‐ECMWF data set are evaluated. In general, local torques show the largest means and variances over mid‐latitudes (especially in the southern ocean), with fluctuations at periods shorter than 1 month dominating the records. In contrast, variability in the globally integrated ocean torque is largest at seasonal time scales (3 months to 1 year); midsouthern and tropical northern latitudes contribute substantially to this variability. Comparison between the global ocean torque and atmospheric angular momentum M indicates that indirect exchanges of momentum between the atmosphere and solid Earth via oceanic stresses cannot be neglected at seasonal time scales. Intraseasonal fluctuations (1–3 months) in both tropical and global ocean torques are coherent with M, but too weak to account alone for the variability in M. At submonthly periods the ocean torque is apparently not important in the atmospheric momentum balance.