WHOI-2002-08

North Brazil Current Rings Experiment:

RAFOS Float Data Report

November 1998 – June 2000

by

Christine M. Wooding, Philip L. Richardson

Marguerite A. Pacheco, Deborah A. Glickson

and David M. Fratantoni

Woods Hole Oceanographic Institution

Woods Hole, Massachusetts 02543

July 2002

Technical Report

Funding was provided by the National Science Foundation through

Grants No. OCE-9729765 and OCE-0136477 to the Woods Hole Oceanographic Institution.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Reproduction in whole or part is permitted for any purposes of the

United States Government. This report should be cited as:

Woods Hole Oceanog. Inst. Tech. Rept., WHOI-02-

Approved for public release; distribution unlimited.

Approved for Distribution:

_____________________________________

Nelson Hogg, Chairman

Department of Physical Oceanography

Abstract

Twenty-one RAFOS floats were tracked at depths of 200-1000 meters in and around several North Brazil Current Rings between November 1998 and June 2000. This was part of an experiment to study the role of these current rings in transporting upper level South Atlantic water across the equatorial-tropical gyre boundary into the North Atlantic subtropical gyre. The float trajectories in combination with surface drifters and satellite imagery reveal the sometimes complex life histories of several rings and their fate as they collide with the Lesser Antilles Islands. This report describes the float trajectories, the velocity, temperature, and depth time series, and a preliminary analysis of the float data.

1. Introduction

North Brazil Current (NBC) rings are large (400 km diameter) anticyclones that pinch off from the NBC retroflection in the western tropical Atlantic near 8N and translate northwestward along the coast of South America toward the Caribbean (Johns et al., 1990; Didden and Schott, 1993; Richardson et al., 1994; Fratantoni et al., 1995). NBC rings have been proposed as one of several important mechanisms for the transport of South Atlantic upper-ocean water across the equatorial-tropical gyre boundary and into the North Atlantic subtropical gyre. Such transport is required to complete the meridional overturning cell in the Atlantic forced by the high-latitude production and southward export of North Atlantic Deep Water. The mechanisms that contribute to NBC ring formation and the structure and dynamics of the rings themselves are not well understood. The purpose of this study is to obtain, for the first time, comprehensive observations of the NBC retroflection, the NBC ring formation process, and the physical structure and properties of NBC rings as they translate northwestward along the low-latitude western boundary. The goal is to understand the process of NBC ring generation and to quantify the role of NBC rings in cross-equatorial and cross-gyre transport within the Atlantic meridional overturning cell. Specific objectives of the float component of the NBC Rings Experiment are to:

1. Measure and describe the physical structure of NBC rings after separation from the NBC, including both near-synoptic and time-evolving temperature and velocity characteristics.
2. Determine characteristic rates of translation, mixing, and decay as NBC rings move northwestward toward the Caribbean Sea.
3. Identify the long-term fate of South Atlantic water trapped within the ring core, and determine the effective intergyre ring transport in several temperature/density watermass classes.

2. Float Deployment

Four cruises were conducted during the North Brazil Current Rings Experiment. On each cruise, shipboard ADCP, XBT, and CTD-LADCP surveys were used to locate NBC rings and to measure their physical properties (Fleurant et al., 2000a, b, c). Subsurface floats were deployed during the first three cruises (see Table 1) and during a non-NBC cruise following cruise 2. Two acoustic sound sources (S1 and S2) were deployed on the first cruise and retrieved on the fourth. An ALFOS float was also launched on the first cruise in order to monitor the sources.

Table 1. Number of RAFOS floats launched during each cruise.

Cruise	Beginning Date	Ending Date	RAFOS Floats Launched
			Total	Into Rings
NBC 1	11-07-98	12-11-98	14	13
NBC 2	02-06-99	03-09-99	11	9
NBC 3	01-29-00	02-24-00	2	2
NBC 4	07-06-00	23-06-00	0	0

Note: All cruises were on the R/V Seward Johnson.

The floats equilibrated, on average, deeper than their target depths. We have divided them into three main depths: a shallow layer near 250 m, a mid-depth layer near 550 m, and a deep layer near 900 m. North Atlantic Deep Water (NADW) is located below roughly 1000 m; it flows southward on average, counter to the translation direction of the rings. The swirl velocity of several rings extended into the NADW, but because one of the goals of the NBC Ring Experiment was to measure the northward flux of South Atlantic Water, the floats were confined to depths less than 1000 m.

3. Description of the Floats

The North Brazil Current Rings Experiment used three different kinds of subsurface floats. One was an ALFOS (ALACE-RAFOS) float, which surfaced once every 10 days and transmitted like a very short-mission RAFOS float. There were six RAFOS floats of the "standard ROM" type, and there were 22 of the latest kind of RAFOS float, the DLD2 (see Appendix A for a discussion of DLD2 technology).

Eighteen DLD2 floats were purchased from Seascan Corporation of Falmouth, Mass., and assembled, calibrated (temperature and pressure), and ballasted at WHOI. Four other DLD2 floats, prototypes provided earlier by Seascan that had been used for testing in the WHOI float laboratory, were also used. Six additional floats were assembled from spare parts and parts of floats retrieved from earlier experiments. The ALFOS float, assembled at WHOI, was used to monitor the sound sources.

The floats recorded temperature, pressure, and times of arrival (TOAs) of sound signals transmitted by moored sound sources. At the end of their missions, the floats dropped ballast weights, rose to the ocean surface and transmitted data to WHOI via the Service Argos Inc. satellite system.

Four of the 22 DLD2 floats were never heard by Service Argos, and one (float 037) was not tracked because it went deeper than its depth limit and surfaced after only 36 hours. Two of the six older-design floats were not tracked because of poor-quality TOAs. Overall we obtained 21 trajectories, although some were shorter than planned (Figure 1). Considering that we purchased 18 floats, this seems like a good record.

The majority of floats was deployed on two cruises roughly two months apart. In order to have most of the floats complete their missions in the same time period, the later-deployed floats were scheduled for shorter missions than the earlier ones. The DLD2 floats were set to repeat their listening schedule every 12 hours. Two of the standard-ROM RAFOS floats were also set to repeat listening every 12 hours; the other four repeated every 24 hours.

Figure 1. Float duration chart. Float numbers are shown to the left of the line that indicates submerged time. The dark part of the line indicates when the float was tracked. Floats are listed in order of launch date.

The floats were generally set to listen starting at 0100, 0130, and 0200. Nothing was gained by including the 0200 window, and little was lost by omitting the 0030 window. However, the tracking program is hard-wired for 0030, 0100, 0130, and for 17 columns per record. This meant that an extra step was needed, splitting the longer records, and shifting all TOAs and correlations right four columns, inserting a "null" 0030 window.

Columns before conversion:

1 | 2 3 4 5 | 6 7 8 9 | 10 11 12 13 | 14 15 16 17

count | 0100 | 0130 | 0200 | T&P

Columns after conversion:

1 | 2 3 4 5 | 6 7 8 9 | 10 11 12 13 | 14 15 16 17

count | 0030 | 0100 | 0130 | T&P

(null)

Neither kind of RAFOS float stored the complete temperature or pressure values. This means that a target value must be assigned, and the appropriate rollover value applied until the result is within range of the target. The DLD2s used the last-recorded pressure and temperature of the mission as a default target (for earlier floats, the target was assigned). These values were often at the maximum or minimum. The floats also experienced a wide range of temperatures (especially those reaching the Caribbean or going ashore). For these reasons, a number of floats (18 of 21) needed second and sometimes third targets to get the pressure and temperature correct, before they could be processed.

The floats were ballasted to drift at a preset depth. This involved weighing the floats in air and in water at different pressures (see Tables 2 and 3). Two floats (037 and 038) surfaced early due to exceeding their pressure limits. On average the floats equilibrated 73 m deeper than their ballast depths. RAFOS floats 431 and 432 were always shallower by 25 to 75 m than their target depths. Floats 156 and 193 had mean pressures close to their targets. The DLD2s were all deeper than their target depths throughout their missions.

Table 2. RAFOS float summary – launch and end positions

	Launch				Surface				Target	Equil	Mean	Mean
Float	Date	Lat	Lon		Date	Lat		Lon	Depth	Depth	Temp	Depth	Ring	CTD
	yymmdd	˚ N	˚ W		yymmdd	˚ N		˚ W	m.	m.	˚ C.	m.
	DLD2 Floats
022	990219	9.179	52.439	000513		12.633	52.287		450	550	7.38	530	R3	2-42
023	990219	9.179	52.439	000513		7.981	47.296		800	910	5.21	907	R3	2-42
024	990219	8.780	52.481	000513		13.063	74.758		200	250	11.90	268	R3	2-41
025	990219	9.179	52.439	000513		15.086	70.016		200	250	13.90	282	R3	2-42
029	981208	8.841	56.242	000530		10.018	77.339		200	275	11.70	269	R1
030	981206	6.405	50.521	000527		8.756	48.164		450	555	7.44	531	R2	1-52
031	990213	10.266	57.203	000507