When people talk about Hawaii, the first thing they think of is probably a beach-side holiday. But for me, I think of the US time station WWVH and the KVM70 radiofax transmissions. Strange, I know … I must get that brain of mine checked out!
That being said, KVM70 can be heard from Australia with some regularity – I’ve posted about receiving their faxes several times in the past but made no concerted effort to catalogue their charts owing to the limited amount of time the signal propagates favourably.
Thanks to publicly available KiwiSDRs, this is no longer an issue, so I went fax-collecting via Northland Radio ZMH292’s KiwiSDR in Bay of Islands, New Zealand and KPH in Point Reyes, California. I did try a number of other stations including AI6VN/KH6 but being “too close” to the signal can be worse than being a little far.
As usual, faxes presented represent a snapshot in time and are not updated. Those interested in up-to-date NWS/NOAA products sent by radiofax can retrieve many of them them over the internet at ftp://tgftp.nws.noaa.gov/fax/Amaster_index.html. The content of the faxes remains copyright of the National Oceanic and Atmospheric Administration of the United States of America and/or the National Weather Service of the United States of America. Faxes are reproduced for documentary purposes to illustrate the technical characteristics of the transmitted charts, the format of the charts themselves and as proof of the activity status, propagation and reception on the dates noted within.
As there are a fair number of different charts, I’ve decided to group the faxes into a few categories for easier viewing. Best reception often occurred on the 11090kHz frequency, although all listed frequencies were seen active at various points within the day.
Test and System-Related Charts
The first chart is a “test chart” sent at 1717UTC on 25th January 2019. This is a rather sparse test chart with mostly white and a CQ call from HFO, which I believe stands for “Honolulu Forecast Office”. Calling CQ by radiofax is definitely not something I’m used to – but all the NWS/NOAA stations seem to do it, although the other “family” stations tend to do it with a nicer fax that has a logo. Interesting that HFO is used rather than their callsign of KVM70.
Key characteristics of the KVM70 transmissions include some data encoded in the first few image lines, manifesting as white dots on black in the top left corner of the image area. The margins of the image area are also black-level resulting in a fax with black borders. Standard length phasing and start/stop tones are provided.
There are some instances when the chart is not available, so a very short fax with the words “CHART NOT AVAILABLE” are sent – this was received 1940UTC on 25th January 2019.
Other times, there are product discontinuations which leave “holes” in the fax schedule. As a result, a fax of very similar dimensions to a chart is sent with a message notifying of the status – note the very “retro” font and the fact it is sent “sideways”. This example received 1724UTC on 25th January 2019.
To aid chart interpretation, a “Symbols & Test Chart” is also sent, which is a nice touch. This example received 1340UTC on 25th January 2019, with an updated date that looks to be 3rd January 2007 (perhaps, or 2002? or 2003? – very hard to ascertain).
The next pair of system related charts are the transmission schedules, sent in two parts, with the above examples receievd 0100 and 0120UTC on 26th January 2019. The schedule appears to have been last updated 4th March 2015 and features relatively bold, legible monospace font. Surprisingly, the format of the schedule differs slightly to that of the other NWS/NOAA family transmissions (as you will see later when I get around to posting their details). They also list that the charts may be found online at http://weather.noaa.gov/fax/hawaii.shtml but this link seems to be broken.
A number of satellite images are disseminated via KVM70 including the Pacific GOES image, the above received from home on 25th December 2018 being a little difficult to receive elsewhere due to propagation. Use of the greyscale abilities of the fax is shown, the above being a relatively “short” fax.
There is also the East Pacific GOES image, received 0635UTC on 25th January 2019, resulting in a taller chart that has “square” grid. It’s nice that this satellite image fills the whole image area.
There is also a South-West Pacific GOES image, received 0649UTC on 25th January 2019. This is the shortest of the satellite images with a slight “blooper” of truncating the latitude marker legends.
Finally, there is the Tropical GOES image, received 2130UTC on 25th January 2019. KVM70 (and other NWS/NOAA family stations) definitely stand out with a total of four different satellite images by fax, an impressive number. This is not unsurprising as the USA is one of the few countries with a mature satellite weather observation program.
Tall Analysis Charts
As so few of the KVM70 charts are tall charts, I decided that the tall “portrait-shaped” charts deserve their own category.
The first example is a Cyclone Danger Chart, received 0535UTC on 26th January 2019. Bold lines clearly denote land mass, however, as there is no active cyclone in the area, the chart is mostly blank. They still send the chart anyway, which is not the normal practice for some other stations.
The next chart is the Pacific Streamline Analysis received 0555UTC on 26th January 2019. This shows a good use of varying thicknesses of lines, with most lines being quite thick to allow for easier interpretation during periods of poor reception. Labelled in small text (which is easy to miss) are the cities of Brisbane, Nadi, Christmas Island, Pitcairn amongst others.
Finally, the last “tall” chart is the Pacific Surface Analysis which covers the same area as the chart above. The above chart is received 0615UTC on 26th January 2019. The sent chart is portrait orientation but needs to be rotated for viewing – the remainder of the charts from KVM70 are “short” charts.
A number of different analysis products are disseminated including the Pacific Surface Analysis which comes in two parts, sent sideways in a short-chart format. The first part is shown above, received 0330UTC on 26th December 2019.
This is followed by the second part of the chart at 0343UTC. This has a slight blooper in it in that the end tone has cut off some of the legend, which was at the top of the fax before but instead resides on the bottom of the second part.
A Tropical Surface Analysis is also provided at 0356UTC, sent the “right way up” but with significant white borders on left and right and a lot of blank space at the bottom. The name of the analyst is also written on the chart, with the logo overlaid over part of the text.
Other forms of analysis offered include “Sea-State Analysis”, the above chart by Rowland, received 0320UTC on 26th January 2019. Quite a short chart, but this appears to be the “default” KVM70 size.
An analagous sea state analysis for the “tropical” area is formatted slightly differently, with significant white borders on either side and the stop-tone cutting into the chart area. This example was forecasted by Cam and was received 0240UTC on 26th January 2019.
Finally, there is the satellite-derived sea surface temperature analysis, with KVM-70 proudly written underneath the logo in the left margin. This chart is relatively unique in its format, unlike most other charts in terms of the logo positioning and the callsign being mentioned. Received 2328UTC on 25th January 2019.
A number of different surface forecast products are available. The Pacific Surface Forecast series of charts is a more “square” shaped chart, the above example is the 24 hour Pacific Surface Forecast received 1901UTC on 25th January 2019. Similar graphic conventions are used with a clear identification of the transmitting entity by logo and text, which is good. The data embedded in the first few lines is longer than most other charts – its meaning is not clear to me but may be used with some special “selective” receivers.
The above example is the 48 hour Pacific Surface Forecast, received 1914UTC.
This is immediately followed by the 72 hour Pacific Surface Forecast received 1927UTC on the same day.
There is also another type of surface forecast chart – the above is a 48 hour Surface Forecast chart of the type that is transmitted in-between other charts and covers a different area – e.g. at 0816 and 2016UTC. The above example was received at home on 25th December 2018.
There is also the “tropical” surface forecast charts which take a different format, including a legend that includes the forecaster’s initials. This chart is bordered by white and does not utilise the full image area, and is a short chart. The above example was received 0200UTC on 26th January 2019 and represents a 24 hour surface forecast.
This was followed by the 48 hour surface forecast at 0210UTC.
And, as expected, a 72 hour surface forecast at 0220UTC on the same day.
Wave and Wind Forecasts
The most basic wind and wave forecast is the 24 hour chart, the above example received 2006UTC on 25th January 2019. Note that while there is another forecast scheduled before with the same title, the chart is replaced with the discontinuation message shown prior. This chart is sent sideways with some margin on both sides.
Which leads to the surprising result that the 48 hour chart follows a different format – this one received 2026UTC. Note that due to the American nature of the station, all wave heights are given in feet.
Wind and wave forecasts for different areas seem to be formatted differently. This looks like it was originally a tall format chart which had been rotated to save on transmission time. The above 48 Hour Wind/Wave Forecast was received 2302UTC on 25th January 2019.
This is followed by the 72 hour forecast following the same format at 2315UTC.
The tropical Wind/Wave Foreceasts all follow a consistent formatting, the above is a 24 hour forecast received 0250UTC on 26th January 2019.
Which is followed by the 48 hour forecast at 0300UTC.
And by the 72 hour forecast at 0310UTC.
Two types of wave period graphs were recorded, including this 72 hour tropical wave period and swell direction chart received 0230UTC on 26th January 2019.
The other wave period/swell direction graph was a 96 hour forecast received 2036UTC on 25th January 2019.
Even the best radiofax stations occasionally have bloopers – a slip in schedule might result in a timing issue that can result in strange things happening.
Received 2116UTC on 25th January 2019, this is a “runt” fax with the ID dots in the top left but only a very small amount of white image content before the stop tone. Shortest fax ever?
Received 2056UTC on 25th January 2019, the fax transmission appears to have been sending the 96h Wind & Wave Forecast but this partial transmission was terminated without a stop tone and then followed by a start tone for the next fax – 24h Wind/Wave Forecast. It could have been the result of a scheduling issue.
Despite this, KVM70 has been a very reliable station with most charts being sent exactly at the time they are scheduled just like clockwork, although noting some products within the schedule may not be available as they may have been discontinued.
I can say with some confidence that the collection above represents practically all the varieties of faxes you might normally encounter from KVM70. There is a variety of chart shapes and sizes although short charts seem the more common variety. It is good to see the use of bold lines, monospaced text and greyscale satellite imagery.
As part of the US NWS/NOAA “family” of radiofacsimile stations including NOJ, NMC, NMF and NMG, they share a lot of commonalities to the point that some charts are available on other transmitters as well. All of the US NWS/NOAA family transmissions are easily identified by the “coded dots” in the upper left of the image area and black margins in the image area.
KVM70 is a fairly nice system to try and tune into because like other NWS/NOAA family transmitters, it runs for a significant portion of the day on two of three allocated frequencies, making it easy to determine if you’re getting any signal at all.