No abstract

[1] The flow through the Yucatan Channel and into the Gulf of Mexico is a major component of the Gulf Stream and the subtropical gyre circulation. Surprisingly, however, little is known about the forcing and physical parameters that affect the current structures in the Channel. This paper attempts to improve our understanding of the flow through the Channel with a detailed analysis of the currents obtained from a primitive-equation model that includes the Gulf and the entire Caribbean Sea and forced by 6-hourly wind from ECMWF. The analysis includes two parts: First, the overall statistics of the model results, including the Loop Current (LC) variability, the frequency of LC eddyshedding, and the means and standard deviations (SD) of transports and currents, are compared with observations. Secondly, an Empirical Orthogonal Function (EOF) analysis attempts to identify the physical parameters responsible for the dominant modal fluctuations in the Channel. The model LC sheds seven eddies in 4 years at irregular time intervals (6.6, 7.1, 5.3, 11.9, 4.2, 10.9 months). The model's upper (thickness $800 m) inflow into the Gulf of Mexico occupies two-thirds of the Channel on the western side, with a near-surface maximum (4-year) mean of around 1.5 m s À1 and SD % 0.4 m s À1 . Three (return) outflow regions are identified, one in the upper layer (thickness $600 m) on the eastern third of the Channel, with mean near the surface of about 0.2 m s À1 and SD % 0.14 m s À1 , and two deep outflow cores, along the western and eastern slopes of the Channel, with (Mean, SD) % (0.17, 0.05) and (0.09, 0.07) m s À1 , respectively. The total modeled Channel transport varies from 16 to 34 Sv (1 Sverdrup = 10 6 m 3 s À1 ) with a mean around 25 Sv. The above velocity and transport values agree quite well with observations by Maul et al. [1985], Ochoa et al. [2001], and Sheinbaum et al. [2002]. The deep return transport below 800 m was found to correlate with changes in the Loop Current extension area, in agreement with the observational analysis by Bunge et al. [2002]. The EOF mode#1 of the along-channel currents contains 50% of the total energy. It is surface-trapped, is 180°out of phase across the channel, and correlates well (correlation coefficient g % 0.8) with the cross-channel vacillations of the LC frontal position. The EOF mode#2 contains 18% of the energy, and its structure mimics that of the mean flow: dominated by two vertically more coherent regions that are 180°out of phase across the Channel. The mode is dominated by two periods, approximately 11 months and 2 months respectively, and correlates (g % 0.7) with the upper-channel inflow transport. The third and fourth modes, together, account for 18% of the total energy. Their combined time series correlates (g % 0.66) with the deep current over the sill, and is dominated by fluctuations with a period %205 days coincident with the dominant low-frequency fluctuations inherent in Maul et al. 's [1985] sill measurement. Thus the dominant mode of flow fluctuations in the Yucatan...

No abstract

No abstract

[1] Variability in sea level at the longest periods observable in modern records has recently been found to be well correlated with local atmospheric pressure. At shorter periods, however, longshore winds are known to be one of the primary mechanisms for forcing sea level variability on the eastern margins of the ocean. There is a remarkable ∼80 mm drop in mean sea level on both the eastern North Pacific and North Atlantic coasts between the late 1800s and early 1900s; it is found here to be in agreement with longshore wind forcing from the equator up to the latitude of the observed tide stations. Better-resolved data beginning in 1960 show that the delay near the annual period between wind forcing and sea level is approximately 1 month. The relative high in sea level in the late 1800s on the west coast of Europe appears to have propagated westward across the Atlantic as a long Rossby wave and then to have been seen on the east coast of the United States. Because many features in long-term sea level variability are correlated with wind forcing on time scales from annual to decades, it will be prudent to base conclusions about long-term sea level rise on the longest records available. The results here are based on forcing by longshore winds; the related issue of the extent to which open ocean wind curl may also be responsible is not addressed here, nor the extent to which coastal sea level observations are representative of the open ocean.