Ralph F. Milliff

Spatiotemporal Hierarchical Bayesian Modeling Tropical Ocean Surface Winds

Spatiotemporal processes are ubiquitous in the environmental and physical sciences. This is certainly true of atmospheric and oceanic processes, which typically exhibit many different scales of spatial and temporal variability. The complexity of these processes and the large number of observation/prediction locations preclude the use of traditional covariance-based spatiotemporal statistical methods. Alternatively, we focus on conditionally specified (i.e., hierarchical) spatiotemporal models. These methods offer several advantages over traditional approaches. Primarily, physical and dynamical constraints can be easily incorporated into the conditional formulation, so that the series of relatively simple yet physically realistic conditional models leads to a much more complicated spatiotemporal covariance structure than can be specified directly. Furthermore, by making use of the sparse structure inherent in the hierarchical approach, as well as multiresolution (wavelet) bases, the models can be computed with very large datasets. This modeling approach was necessitated by a scientifically meaningful problem in the geosciences. Satellite-derived wind estimates have high spatial resolution but limited global coverage. In contrast, wind fields provided by the major weather centers provide complete coverage but have low spatial resolution. The goal is to combine these data in a manner that incorporates the space-time dynamics inherent in the surface wind field. This is an essential task to enable meteorological research, because no complete high-resolution surface wind datasets exist over the world oceans. High-resolution datasets of this type are crucial for improving our understanding of global air–sea interactions affecting climate and tropical disturbances, and for driving large-scale ocean circulation models.

Deep convection in the Irminger Sea forced by the Greenland tip jet.

Open-ocean deep convection, one of the processes by which deep waters of the worlds oceans are formed, is restricted to a small number of locations (for example, the Mediterranean and Labrador seas). Recently, the southwest Irminger Sea has been suggested as an additional location for open-ocean deep convection. The deep water formed in the Irminger Sea has the characteristic temperature and salinity of the water mass that fills the mid-depth North Atlantic Ocean, which had been believed to be formed entirely in the Labrador basin. Here we show that the most likely cause of the convection in the Irminger Sea is a low-level atmospheric jet known as the Greenland tip jet, which forms periodically in the lee of Cape Farewell, Greenland, and is associated with elevated heat flux and strong wind stress curl. Using a history of tip-jet events derived from meteorological land station data and a regional oceanic numerical model, we demonstrate that deep convection can occur in this region when the North Atlantic Oscillation Index is high, which is consistent with observations. This mechanism of convection in the Irminger Sea differs significantly from those known to operate in the Labrador and Mediterranean seas.

Wind Stress Curl and Wind Stress Divergence Biases from Rain Effects on QSCAT Surface Wind Retrievals

Abstract Surface vector wind datasets from scatterometers provide essential high-resolution surface forcing information for analyses and models of global atmosphere–ocean processes affecting weather and climate. The importance of realistic amplitude, high-wavenumber, surface wind forcing from scatterometer data has been demonstrated in a variety of ocean modeling applications. However, the radar backscatter signal from which surface vector wind estimates are retrieved is attenuated and/or contaminated in heavy rain. The QuikSCAT (QSCAT) dataset flags rain-contaminated wind vector cells where retrievals are either highly uncertain or not available. Zonal and annual averages of wind stress curl and divergence for 2000, 2001, and 2002 are derived and compared across three surface wind datasets: QSCAT only, reanalysis winds from the National Centers for Environmental Prediction (NCEP reanalysis), and blended QSCAT+NCEP. Missing QSCAT surface wind retrievals due to rain contamination lead to statistically sign...

Ocean general circulation model sensitivity to forcing from scatterometer winds

Enhanced global surface wind fields are constructed from a blend of NASA scatterometer (NSCAT) and ERS 2 scatterometer data and National Centers for Environmental Prediction (NCEP) analyses, at 6-hour intervals, for a repeatable annual cycle from August 1996 through July 1997. Wind field properties (wind speed, zonal and meridional wind stresses, wind stress curl, and kinetic energy input) for the enhanced winds are compared with the NCEP analyses for the same time period. Large-scale, zonal patterns dominate annual average difference maps for wind speed and both components of wind stress. Wind stress curl differences are largest in the subpolar North Atlantic and in the midlatitudes of the southern hemisphere. The importance of wind field differences are measured by their impacts on the response of an ocean general circulation model (OGCM). Twin experiments are performed using the National Center for Atmospheric Research Climate System Model ocean component in stand-alone mode. The annual mean OGCM responses to the enhanced winds and to surface winds from the NCEP analyses are compared for barotropic stream function, surface velocities, upper ocean upwelling, sea surface temperature (SST), surface heat flux, meridional overturning stream function, and total northward heat transport. Differences in the annual mean responses are attributable to differences in the mean forcing and not to the mesoscale signal that is present in the enhanced winds but not resolved by the OGCM. Differences in SST and surface heat flux are partitioned between local thermodynamic balances and balances involving ocean dynamics as well. Northward heat transport differences of order −0.2 PW in the southern hemisphere midlatitudes are consistent with weaker eastward and southward stresses in the westerlies and a 27% reduction in kinetic energy input from enhanced winds. This difference in northward heat transport is compensated in the southern hemisphere tropical Pacific where surface stresses and the OGCM surface response to enhanced winds are more westward and more divergent, resulting in 27% greater kinetic energy input. An extensive appendix details the realistic high-wavenumber character of the enhanced winds.

Basin-Scale, High-Wavenumber Sea Surface Wind Fields from a Multiresolution Analysis of Scatterometer Data

Abstract A numerical technique sensitive to both spectral and spatial aspects of sea surface wind measurements is introduced to transform the irregularly sampled satellite-based scatterometer data into regularly gridded wind fields. To capture the prevailing wavenumber characteristics (power-law dependence) of sea surface wind vector components, wavelet coefficients are computed from the scatterometer measurements along the satellite tracks. The statistics of the wavelet coefficients are then used to simulate high-resolution wind components over the off-track regions where scatterometer data are not available. Using this technique, daily wind fields with controlled spectral features have been produced by combining the low-wavenumber wind fields from ECMWF analyses with the high-wavenumber measurements from the ERS-1 scatterometer. The resulting surface wind fields thus reflect nearly all available measurements affecting surface wind, including the synoptic surface pressure. The new surface wind forces a b...

The Existence and Vertical Structure of Fast, Eastward-Moving Disturbances in the Equatorial Troposphere

Abstract Eastward phase propagation, at speed faster than 30 m s−1, of a signal in the equatorial troposphere of the Eastern Pacific is detected, first in historical meteorological observations and then in more recent data. A first baroclinic mode vertical structure is identified with this signal in separate analyses based on linear theory and complex empirical orthogonal functions, respectively. This rapid, eastward signal is conceptualized as a far-field dispersion product of strong convection associated with the intraseasonal tropical oscillation in the Indian Ocean and Western Pacific.

The Global Distribution of the Time-Average Wind Stress Curl from NSCAT

Abstract The time-average wind stress curl field for the global ocean is computed from the wind retrievals of the NASA Scatterometer (NSCAT) mission spanning the period 1 October 1996–29 June 1997. Particular attention is paid to large-amplitude, small-scale “patchiness” in the average wind stress curl over the major ocean basins, and to long and narrow wind stress curl features that occur along ocean eastern boundary regions. The 9-month-average wind stress curl field from NSCAT is examined at 0.5°, 1°, and on a Gaussian grid consistent with T62 truncation in a spectral forecast model. The latter field is compared with the average wind stress curl field from NCEP analyses for the same period. Artifacts in the NCEP average overlap the regions of boundary wind stress curl extrema in the high-resolution averages from NSCAT. The artifacts are attributed to the effects of spectral truncation and tall near-coastal topography in the NCEP forecast model. Possible explanations are discussed for the boundary wind ...

Composite life cycle of maritime tropical mesoscale convective systems in scatterometer and microwave satellite observations

This study examines scatterometer-observed surface wind divergence and vorticity, along with precipitable water (PW), across the life cycle of tropical maritime mesoscale convective systems (MCSs) as resolved in 0.5° data. Simple composites were constructed around first appearances of cold (210 K) cloud tops in infrared (IR) data at 3-hourly resolution. Many thousands of such events from the tropical IndoPacific in 2000 were used. Composites of subpopulations were also constructed by subdividing the dataset according to IR event size and duration, as well as by prevailing values of PW and vorticity at a 5° scale. The composite MCS life cycle here spans about a day and covers a few hundred kilometers, with a remarkable sameness across subpopulations. Surface wind convergence and PW buildup lead cold cloud appearance by many hours. Afterward there are many hours of divergence, indicative of downdrafts. Contrary to motivating hypotheses, the strength of this divergence relative to convergence is scarcely different in humid and dry subpopulation composites. Normalized time series of composite vorticity show an evolution that seems consistent with vortex stretching by this convergence–divergence cycle, with peak vorticity near the end of the period of convergence (3 h prior to cold cloud appearance). In rotating conditions, the common 1-day MCS life cycle is superposed on large-scale mean vorticity and convergence, approximately in proportion, which appear to be well scale-separated (covering the whole of the 48-h and 5°–10° averages) and are as strong as or stronger than the MCS signature.

The General Circulation Responses of High-Resolution North Atlantic Ocean Models to Synthetic Scatterometer Winds

Abstract High-resolution (1/5°×1/6°) quasigeostrophic models of the North Atlantic Ocean are forced by daily wind stress curl fields of controlled wavenumber content. In the low-wavenumber case, the wind stress curl is derived from a low-pass filtering of ECMWF wind fields such that the retained wavenumber band is observed to obey a k−2 power law in the spectrum for each day (where k is the wavenumber vector). In a second case, the wavenumber content of the wind stress curl fields is comparable to that derivable from an ideal scatterometer-wind dataset. Decadal-average streamfunction fields are compared with a climatology of dynamic topography and compared between the model calculations driven by these synthetic wind stress curl datasets with the goal of testing the sensitivity of the general circulation to high-wavenumber forcing. The largest signal in decadal-average streamfunction response to high-wavenumber forcing occurs in the eastern basin of the North Atlantic. Fields of mean kinetic and eddy kine...

The Evolution of Boundary Pressure in Ocean Basins

Abstract The boundary pressure adjustment process on an ocean basin scale is elucidated in two sets of numerical experiments. First, an initial-value problem is posed in a primitive equation shallow-water model that leads to significant changes in the pressure averaged along the boundary in a closed rectangular ocean basin. These results are compared with the analogous problem in a shallow-water quasigeostrophic model where the boundary pressure adjustment is parameterized by a consistency constraint that closes the mass, circulation, and energy balances in quasigeostrophy. There is very good agreement in the evolution of the boundary-average pressure and qualitative agreement in the evolution of the balanced motions in the interior. Second, idealized Kelvin wave experiments are posed in the primitive equation system on an f plane, a β plane, and a β plane in a domain of doubled dimensions. For β≠0, a scattering process is evident as the initial Kelvin wave transits the first meridional boundary it encoun...