More late changes to gas explosion and noise studies as Karpowership enters final lap

Karpowership consultants have made further last-minute changes to an environmental impact assessment (EIA) study, the effect of which is to reduce the predicted negative effects of the Turkish floating powerships plan.

Some alterations came to light last month when Our Burning Planet uncovered several inconsistencies between a new fire and explosions risk study for the powerships and a previous version published more than a year ago. These changes have largely been attributed to “software updates” in the computer models used to predict death and casualty risks should a major gas fire or explosion break out on Karpowership vessels anchored in the ports of Richards Bay, Ngqura or Saldanha Bay.

These belated explosion risk revisions come at a time of heightened public awareness around gas explosions following the devastating Boksburg disaster that killed 37 people after an LPG road tanker explosion on Christmas Eve.

Now, other observers have also picked up further revisions to some of the latest EIA studies – including “dramatic” reductions in the noise levels predicted to be generated by powerships over the next 20 years.

Some of these late changes to the original EIA noise studies have been attributed to “new information” emerging from Karpowership and other sources about ship structure and the silencing equipment fitted to the Turkish vessels. 

Read in Daily Maverick: “Karpowership’s environmental hazards controversially reduced in new EIA specs”

After years of controversy, a final EIA report is due to be published later this month and then resubmitted for approval to the Department of Environmental Affairs, which rejected the first Karpowership EIA in 2021 on the basis that it was flawed and inadequate in several respects.

Last month, Our Burning Planet undertook a detailed line-by-line comparison of a major hazard risk assessment by Karpowership’s Cape Town risk consultant Claude Thackwray. Significantly, we found numerous changes to the studies he published in February 2021 and November 2022.

In his revised study, Thackwray has reduced the predicted fatality distance of a transfer hose flash fire from 879m to 499m. Fatal impacts due to a jet fire dropped from 568m to 188m, while in another Karpowership gas fire scenario, the lethal impact distance dropped from 220m to just 15m.

At the time, Thackwray attributed these changes to software updates to the computer risk modelling tools he uses. But we noticed that many of the changes applied only to Richards Bay and Ngqura – whereas the fatality distance data for Saldanha Bay remained unchanged.

Now, in response to written requests to explain why the software updates did not appear to affect Saldanha, Karpowership EIA consultant Hantie Plomp has provided further responses after consulting Thackwray.



According to Plomp, the specialist (Thackwray) forgot to change the modelled inputs for Saldanha when he compiled the November 2022 report.

“This was an oversight and has been rectified in the (final) report,” she said, also acknowledging a further “typo error” in describing some of the gas explosion risk contours.

This error had also been corrected, she said, asserting that the mistake did not change the overall results or conclusions of the final report.

The acknowledgement of these omissions and belated changes to gas explosion risk predictions in three national harbours come in the wake of the recent Boksburg gas disaster.

While the Boksburg explosion was caused by propane-based liquid petroleum gas (LPG) – not the methane-based liquid natural gas (LNG) that would be burned by Karpowership – there have nevertheless been numerous fatal natural gas explosions in several parts of the world dating back to the early 1940s.

In one of the worst such incidents on record, at least 124 people were killed after an LNG explosion in Cleveland, US, in October 1944.

In January 2004, 27 people were killed and more than 80 injured after an explosion at an LNG facility in Skikda, Algeria. Six months later, another 24 people were killed and more than 130 injured after an LNG pipeline rupture near Ghislenghien, Belgium.

Read in Daily Maverick: “Karpowership gas leviathan bubbles back up — again”

In his report, Thackwray acknowledges the potential for several LNG accident scenarios, including vapour cloud explosions that could cause “extensive property damage and injury or loss of life”.

Overall, however, Thackwray suggests that Karpowership LNG operations in South African  ports “pose a very low risk” – comparable to “ordinary gas connections in a domestic home” or the risk of people getting killed by lightning.



But the recent changes to the gas explosion risks are not the only EIA revisions to have raised eyebrows in several quarters.

In written comments submitted to Plomp and her colleagues, two non-government environmental groups have expressed separate concerns around last-minute revisions to computer-modelled noise predictions emanating from powerships in Saldanha Bay.

The Green Connection and Natural Justice note that noise results modelled during the first EIA had now dropped “dramatically” in the second EIA.

2022 revised noise study Saldanha

2021 noise study Saldanha.

A commuter bus powered by methane-based compressed natural gas (CNG) burns in the Stockholm city centre in March 2019. The bus – fortunately carrying no passengers at the time – struck the top of a road tunnel and exploded. (Image source: Twitter/ Popular Mechanics/ Aftonbladet)