CAUTIONARY NOTE: BoM 2020 RIP predictions

Packo's prediction for the slack water times at Port Phillip Heads.

Moderator: packo

DAE avatar
packo
Forum Moderator
Posts: 40
Joined: Thu, 22 Mar 2018 8:34 am

CAUTIONARY NOTE: BoM 2020 RIP predictions

Postby packo » Fri, 15 Nov 2019 9:40 am

****** Background ******

Update 15/11/2019: Some weeks back I completed work on the "Packo Predictions" of 2020 slack water times and other info for Port Phillip Heads. Part of that work involves a cross check of "Packo times" with the "Cardno times" that the BoM and other government agencies publish. Due to the quite different approaches used by the two algorithms, the times can differ by up to 20 minutes. Anything more indicates some sort of processing blunder.

Initially this "sanity check" was aimed at detecting any blunders I may have made. Interestingly so far it has only worked the other way around, and I recently identified a problem with some of BoM's "2020 Rip predictions".

(Note that I continue to claim that the smaller <20 minute differences in the Packo times represent improved predictions but because of weather induced variations I can't prove this with sufficiently high certainty at this time.)


****** Notes on BoM's 2020 Rip prediction problem *****

1) Please note the problem concerns only the 2020 Rip predictions on the BoM's "tides webpage", and

2) the accompanying 2020 Rip PDF file that is linked to that webpage.

3) After my notification the National Tide Centre fixed it and updated the BoM webpage on 14/11/2019.

4) Only information sourced between their first release ~March 2019 and 14/11/2019 will still contain errors.

5) On the BOM 2020 Rip tidal streams webpage, where converting times of slacks and max rates to Daylight
Savings or ("local time") involved the new time "rolling over to a new day", then 1 hour was added TWICE.

6) In the 2020 Rip PDF file, the times of slacks and max stream rates are all correctly converted BUT the same
software flaw meant any max stream rate value affected by a day rollover when converted to Daylight Savings
time did not print, leaving just a blank after the time where the rate should be. It is then conceivable that a
max rate time standing by itself may be taken as a slightly out of position "slack time" (ouch!).


******* Who Needs to Do What? *******

Because the issue only effects events in the midnight - 2am time slot, it is perhaps more relevant to yachtsmen (who seem to transit the Rip at all odd hours), rather than divers. However in younger days I do remember nice warm summer nights a few days before a full moon were great times for a late night dive at or near the Heads!

Just to be sure, all divers and any sailing friends that may have either downloaded the "2020 Rip PDF", or who may have cut and pasted any information from the 2020 Rip tidal streams webpage BEFORE 14/11/2019 should DISCARD it and start over.


****** Other Ways the Bad Info Might Persist ******

Cardno Pty Ltd compute slack water & max stream rate times in Eastern Standard Time (UTC+10) because that's what commercial shipping operations use. For some years recreational folk, who mostly observe Daylight Savings Times (UTC+11 between 1st Sunday in October and 1st Sunday the following April) had to remember to add one hour when "daylight savings" was in operation.

A few years back BoM decided that its website tide data etc was best presented with Daylight Savings corrections already applied for those states that observe it. I think this was a sensible move for general public use. Note that the "Victorian Tide Tables", published by the VRCA both as hard-copy and PDF versions, use UTC+10 times and so are NOT AFFECTED by this problem. (but remember to add that hour when needed!)

Other sources of "Rip predictions" such as local newspapers, various magazines, calendar publishers and the like who publish the tables as a weekly, monthly, or yearly block, may choose either timing but with the appropriate warnings. Depending on their printing deadlines and the source of their information, there is a slight chance some of these publications in 2020 may still contain the bad info.


***** How Can You Check if the Source of Rip Times You Will Use in 2020 is Affected? *****

If your "tables" don't use Daylight Savings or "local time" predictions, you can stop right here - the times are OK.

Otherwise check the day's first Max Stream Rate for 5th Jan 2020. The time should read 12:36am or 0036hrs and it should have a rate number of -3.74 knots. If either the time is wrong (ie. 01:36am) or the Rate number is missing then your data source will have quite a number of other errors so consider an alternative.

(Note if your tables don't show this max rate at 12:36am, but it appears at 11:36pm on 4th Jan 2020, then your tables are in UTC+10 times with no daylight savings corrections made - so there should be no problem but you will have to do your own corrections.)

Note that if you use the "Packo Predictions" all times are properly converted Daylight Savings times.

Note the "Packo Tables" don't show the Max Rate times, but rather give the Max Rate and a time range when the current is above 75% of the Max Rate. The Max rate and time band for 5/01/2020 will be shown as the last entry of the previous day. This shows a Max Rate of -3.05 knots and the >75% time band of 23:06 (4th Jan) --> 02:18 (5th Jan). The center of this time band is close to the corrected BoM time of 12:36am (5th Jan).

Note also that the Max Rates for the Packo predictions are a little different than for the BoM predictions because the later "switches" locations:- Rip Bank for Ebb Streams, and Nepean Bank for Flood Streams, ie. at quite different points along the main channel leads. One effect of this switch is to bias the Max Rates towards higher values for Ebb streams because outflow through Rip Bank is concentrated into more of a "jet" than for inflow across Nepean Bank. Compared to the actual volume flow through the Heads their numbers overstate ebb stream strengths and understate flood stream strengths.

While sailors should use the BoM max rates I am sticking to my numbers because divers (who don't dive at high stream times) tend to use these rate numbers as general "flow strength" indicators either for locations well inside the Bay, or for assessing if drift dives are safe or not. I have some concern that the underestimate of flood strengths from the BoM numbers may be enticing some divers into some "flood drift dives" when perhaps they shouldn't.


Update 13/12/2019: Well not many weeks have rolled past since the above post. The immediately above point about flood "drift" dives might have been reinforced by these comments on the VSAG facebook page regarding a December 8th flood drift dive on the Lonnie Wall:-
"We started with a drift starting at North Wall Corner. With the drift line in we could tell that the 2.6 Knot current forecasted was a pipedream. Once in we took an adrenalin filled fast drift. I went for an hour and finished up on Boarfish Reef, you can tell it fast when the kelp is flat and stopping for pics just wasn't an option."

The BoM predictions for that flood stream showed a max of +2.65 knots @ 8:02am, but the Packo prediction tables had it quit a bit faster at +3.08 knots. Also worthy of note is that the force the rushing water exerts goes up as the square of its speed.

cheers,
packo



Who is online

Users browsing this forum: No registered users and 1 guest