NOAA satellite analysis confirms skipper’s account of the knockdown off Brazil that doomed the tall ship
NOAA satellite imagery and meteorological analysis support the captain’s account that a violent downdraft — a microburst — was likely responsible for the Feb. 17 sinking of the four-masted barque Concordia. Ken Pryor, a researcher at NOAA’s Center for Satellite Applications and Research in Camp Springs, Maryland, reviewed infrared satellite data and concluded that powerful convective activity and a localized downburst struck the vessel about 290 nautical miles south-southeast of Rio de Janeiro, Brazil.
Pryor’s analysis identified classic markers of severe convective storms: strong updrafts with overshooting tops — dome-like features that indicate very strong thunderstorm tops — and a “dry-air notch,” where dry air is drawn into the storm. That dry-air entrainment promotes evaporation of precipitation, which cools and accelerates air downward, producing intense, localized downdrafts. According to Pryor, these conditions were present near Concordia roughly 13 minutes before she capsized.
Captain William Curry of the 188-foot, three-masted Canadian sail training ship described abrupt, erratic wind behavior consistent with a sudden downdraft. In his account he said that when a concentrated rain cell passed over Concordia the wind initially increased to about 25 knots and the ship continued under sail. Then the wind backed sharply, the helmsman attempted to bear away, and the wind soon veered onto the starboard quarter while the anemometer showed rapidly increasing gusts. The watch officer lost track of the instrument as wind climbed past 30 knots while the ship heeled rapidly to port and then went over to an angle between 90 and 100 degrees.
The knockdown occurred quickly and violently. Concordia sank within thirty minutes of the event; two merchant vessels rescued all 64 students and crew from life rafts about 40 hours later. Captain Curry reported that the gust tore the storm-reefed mizzen in at least two places and shredded the upper square topsail. He noted the gust that caused the capsize lasted only minutes.
Before the microburst, prevailing conditions were moderate: winds of 16 to 20 knots and seas of 4 to 6 feet. The mate on watch observed a squall cell approaching and reported a dense, roughly one-mile-diameter rain core but no obvious whitecapping ahead of the line. Concordia was already under reduced canvas — only about 40 percent of her sails set, including storm-reefed mizzen, mainsail, main staysail, fore staysail, inner jib, and upper and lower square topsails.
If a powerful downdraft struck almost directly overhead, it would help explain why the ship could not right itself after heeling to deck-edge immersion, Pryor says. Captain Curry, while reserving final determination to the official investigation, agrees that a downward-directed gust of 40-plus knots on a vessel already heeled to an extreme angle presents a very different and far more dangerous scenario than the Concordia had previously encountered over many thousands of safe miles at sea.
Pryor emphasizes that his findings are a reasoned hypothesis based on satellite infrared signatures and the crew’s observations. His technique looks at differences in water vapor and cloud-top temperatures measured by geostationary infrared sensors to identify storm structures and dry-air notches indicative of downburst potential. In other cases where microbursts were suspected — notably the 2004 Lady D capsize in Baltimore Harbor — radar corroborated satellite indications of a convective system with a dry-air notch and a strong localized downdraft.
The Lady D, a pontoon water taxi, capsized in a severe thunderstorm in March 2004, killing five people. Pryor’s reconstruction for that event used both radar and satellite data to show a similar pattern of convective activity and a downdraft that likely produced surface winds in the mid-30s to mid-40s knots. He notes that while popular descriptions sometimes cite 100-knot microbursts, the vast majority of downbursts observed at the surface fall in the 35 to 50-knot range, which is sufficient to overwhelm small or partially canvassed vessels and create sudden, severe hazards.
Pryor’s interpretation does not claim absolute proof; rather, it shows a coherent picture in which satellite-observed storm features, the timing and location of the convective cell, and the crew’s wind and damage reports are all consistent with a microburst-induced knockdown. Radar confirmation is not cited for the Concordia case in his initial analysis, but he notes that satellite signatures commonly align with radar-detected structures in other well-documented microburst incidents.
Following the loss, some attention turned to Concordia’s stability. Classes Afloat, operator of the sail training program, reported that the ship had passed construction stability testing that demonstrated recovery from a 110-degree knockdown. The Transportation Safety Board of Canada and the Barbados Maritime Ship Registry (which had registered the Canada-based vessel) have continued formal investigations into the sinking to determine all contributing factors.
For now, NOAA’s satellite-based findings provide strong meteorological support for the skipper’s account: a sudden, localized downdraft within a convective storm appears to have produced a surface wind burst that the Concordia could not survive. The incident highlights how compact, intense storm processes like microbursts can create extreme, short-duration hazards at sea that challenge even experienced crews and well-found ships.
This article originally appeared in the June 2010 issue.