A fraction too much friction
When hitting the highway chasing waves, how often do you check the local Automatic Weather Station (AWS) for a real-time update on the wind?
If it's your habit - and it should be! - you may have noticed that sometimes the AWS reads offshore yet the water surface isn't as clean and groomed as you'd expect. Instead, the local wind isn't blowing from the exact same direction the AWS is reporting.
Your next question should be: Where is the AWS located? And if it's inland, the mystery is as good as solved.
The reason for the differing directions is friction, and the more inland you head, the more the wind is steered away from the direction experienced immediately on the coast.
Starting from first principles, air flows from high pressure to low pressure. Much like having two water tanks sitting next to each other but with different water levels. If you connect them, water will flow from the tank with the higher water level and pressure to the other tank until they become equal in level. Another example; a punctured bike tube will see the high pressure air from inside flow out of the hole, into an area of lower pressure outside the tube as it tries to equalise.
This is all well and good, but because the Earth rotates, air doesn't simply flow from a high to low in a straight line. The Earth's rotation introduces another force on moving air, known as the Coriolis Force. The Coriolis Force is sometimes called an apparent force, yet it's very real in terms of its applications in our frame of reference, that being the spinning Earth.
What the Coriolis Force does is steer the air parcel flowing from high to low pressure to the left in the Southern Hemisphere (and the right in the Northern Hemisphere). This is applied continuously over time, so the small deflections to the left continue until the magnitude of the Coriolis Force matches that of the Pressure Gradient Force and we achieve Geostrophic Flow. Geostrophic Flow follows the isobars drawn on a weather map, clockwise around low pressure systems in the Southern Hemisphere and anti-clockwise around high pressure systems.
As the Pressure Gradient Force between a high and low increase (think of a deeper low or a stronger high), the air parcels flowing between them accelerate and with all things being equal, the Coriolis Force also increases. This then results in stronger gradient winds flowing parallel with the isobars.
Pressure Gradient Force (grey) plus Coriolis Force (black) results in Geostrophic Flow (blue) along isobars
We then introduce friction.
As wind blows, it experiences friction and this differs depending on the Earth's surface. If blowing over an ocean or lake the friction is less than when blowing over land, and mountains produce even more friction again.
Adding this frictional component into the mix, we then get a slowing of the air parcel's movement and with this a reduction of the Coriolis Force. This then means the Pressure Gradient Force flowing from high to low is greater than the Coriolis Force, steering the wind across isobars and in towards the centre of low pressure systems, and accordingly out of highs.
The angle in general is about 30° off the parallel-running isobars, but even more when moving over rougher surface and land features.
And this is where things become interesting.
Visualise a low pressure system sitting just offshore from the East Coast, producing gale-force southerly winds along the coast. With the steering of the wind into the centre of the low, plus the added frictional effects of the land we'll see south-west winds at the coastal interface, but more inland they'll swing even more westerly in direction. This would see an inland AWS station reading westerly, yet on the coast winds they'll be more south-west, hence adding small bumps and ridges to spots that under a westerly would be clean.
Below is a classic example from last week.
Overlaying the wind forecasts (blue arrows) on top the Mean Sea Level Pressure charts, you can easily see the frictional effects of the land mass, swinging winds 45° from the synoptic and isobaric flow (white arrows), most evident on the southern NSW coast, while out to sea winds follow the isobars more closely.
Looking futher a little further inland, across the Great Dividing Range which provides the most friction, winds are 90° off angle to the the isobars.
During the mornings, terrestrial land breeze effects from overnight cooling will steer the wind more locally offshore, so this becomes less of an issue. Yet as the land breeze weakens through the morning, the effect will be exacerbated. Therefore, when checking AWS observations that aren't coastal it's worth keeping in mind that on the East Coast the winds at the beach are likely to be more southerly than what the inland station is showing. This effect decreases the closer the AWS is positioned to the coast.
For Victoria the same also applies, but when looking at Western Australia, where high pressure systems bring those favourable easterly winds, the wind is also likely to be a touch more south compared to the inland observations as frictional effects turn the wind out of the high. For example, an inland easterly breeze is likely to be more towards south-east at the coast.
The take home messages are thus. Firstly, try to source coastal wind observations for a more accurate representation of the actual winds on the coast, and secondly, if you can only access inland AWS readings then factor in friction.
Comments
Great simple explanation Craig. Reminds me of "Surf Science" by Tony Butt for easy to use explanations.
When checking real time winds, usually to catch the offshore early (SW, WA), I generally check the wind history for last 3 or so hours - if its been onshore all night, no good if it turned offshore shortly before dawn - need to give it a bit of time to clean up!
Whoa cool! This is a cracker. Thanks Craig. I always check local airport readings here a few times a day and have come to the conclusion there is some element of variability from it's reading, 5km inland to the coastal front, but never understood the science behind it. And i also learnt a new word today. Geostrophic. Will be pulling that one out next carpark convo!! haha
A couple of seemingly 'close to the coast' AWSs (i.e. Ballina, Port Macquarie, Tewantin) don't equate to coastal winds very well at all.
Other locations have more complex terrain (i.e. Aireys Inlet), and/or are surrounded by deep water harbours (i.e. North Head), which also has an effect on wind direction and strength.
I'm guessing it's obvious, but is that primarily because they have relatively featureless coastal topography, generally speaking as opposed to somewhere like down South with high cliffs, and much more dynamic coastal features? Or something else.
The immediate topography near the beach at Port Macquarie is pretty hilly. The weather station, located some distance away in a swamp would not record the winds accurately at the beach because of this.
The Port Macquarie weather station unfortunately got moved to the airport some years ago. As the crow flies, the airport is approximately 10k's away, situated in a low lying swamp. Previously the weather station was in the front yard of a house across the road from the beach. Hence the big difference.
Nice one Ben, Surf Uni essentials.
Coriolis make my head spin...
even stranger sometimes in times of weak synoptic winds when you would expect the strongest land breeze effect to be at the coast/sea interface.
fark. awesome read. gonna have to re-read a couple times but. I feel like this applies to my location alot
Good read, thanks.
Pedant mode on - Not sure the water tank is a good example though. The air pressure and water pressure are higher at the bottom of the hill. The water will flow down because of gravity. If a tap is opened then it flows out because the air pressure outside is far lower than the water pressure in the tank.
Pedant mode off.
Ah yes, got me. Have amended.
Good stuff Craig. Had flashbacks to 3 unit geography where I wondered why the isobars direction never quite matched surface level observations. Surface friction. The same sort of friction that causes waves to form yes I see.
The BOM Observation page is my one of my fave pages. It's always fascinating checking the observations of a couple of relatively close together locations and seeing how much they can sometimes vary. I have also noticed early in the morning that when the wind is SE at Murrurundi Gap, and it may be westerly at the time, the winds at the beach will be around to the South before lunchtime. Fun and useless fact: The wind almost never stops blowing at the Murrurundi Gap.
http://www.bom.gov.au/nsw/observations/nswall.shtml?ref=hdr
Thanks Craig - Great article. Do the Swellnet forecasts take into account the effects of the morning land breeze? Or is it just the synoptic setup and you need to compensate for the morning land breeze effect (which I do - more out of optimism then anything else). Also - curious to know if there is a rule of thumb to determine when the land breeze over powers the synoptic wind. In winter, I work on anything that’s forecast as south easterly and less than 10knots going westerly (offshore) in the morning. This approach has failed and resulted in misery many times. Is there a more scientific approach? I’m in South East Qld - I appreciate our winds can be a lot flakier than other parts of Oz.
Offshores are temperature based. Differential between overnight minimums and sea surface temps. Obviously depends on forecast winds speed and direction also but generally if lighter winds are forecast the offshore will kick in overnight once the land temps drop 4 deg below the sea surface temp. Which explains why Spring is so shit for early morning offshores. Warmer overnight land temps and coldest water temps. Opposite in Autumn when the water is the warmest.
Thanks mate - the 4 degree differential makes sense - good to put a number to a decision I’ve always had a rough guess at. Also - I’ve never taken the time to work out why Autumn is way in front of Spring for offshores. There it is - thanks heaps for the response.
Sweet. Didn't know that either. Still a novice. Cheers Don!
Yeah westof... I note the difference between the readings at Nowra and Kiama, only 50km or so apart, are often up to 45 degrees different, sometimes 90deg even when there isn’t a southerly change. (eg W at Kiama, SSW at Nowra)
Not sure how much topography affects that as well - there’s a prominent spurline that comes off the escarpment just south of Kiama that regularly seems to have different winds N and S of it.
Yep for sure, and it's good for tracking a southerly change as it moves up the coast, to help fine tune your window of opportunity on some particular day.
I’ve used that to my advantage on many an occasion at a southern end point, esp during summer where the NNE wind will be messing up a decent NE windswell but there’s a southerly coming up the coast. Always seems to be a 30-60 minute window before the punters get on it.
Interesting article wondering what role the friction aspect has on Syd northern beaches, which often appear to hold out against a southerly a bit longer than the ES beaches to the South and even the cenny coast at times. Or is this location more down to the localised impact of the harbour to the immediate south creating an outflow that steers the wind around?
Not entirely sure on the mechanics, but it's very likely related to drainage patterns through Sydney Harbour and the Sydney basin. The Pittwater may also have an influence too.
Great article Ben. In Perth, we often see a significant differential in the arvo seabreeze between Rockingham (circa 80kms south) to the metro wind direction, sometimes 30-60 degrees out. Given the seabreeze here (as I understand it) is primarily driven by the in-land land mass heating through the day to create a semi-localised high pressure zone, I'd always kind of attributed the difference between perth and Rocko to be an urban 'heat island' effect changing the wind vector along the metro coast, thoughts on this contributing and to what degree?
Actually, Craig did the article!
Re: differences in Rockingham/Perth sea breeze direction, I'm really not too sure. There've been a few studies done on the Freo Doctor so I'll see if I can find any relevant info.
However there's only one AWS between Bunbury and Fremantle - at Mandurah - and its siting near the marina - has always resulted in lower wind speeds from the southern quadrant IMO.
I also wonder if Collins Pool (is that the right name?) and has an influence too?
Ah my apologies Craig!
I'm going off my wind checks on Seabreeze each afternoon in summer, I'm fairly sure seabreeze readings from random /small / private weather stations (?) I suspect the quality of data of these would be less than the AWS you mention above, so might also be a factor.
I've just had a look at Collins Pool aka Dawesville cut area, and realised that the surrounding area to the east is heavily forested with a small escarpment, where as land north and east of the city is more barron, non forested and i think is nominally flat. So perhaps the diff between metro and rocko could be mostly attributed to friction (forest and escarpment vs urban / flat) and location of the low pressure area -i.e. the forested areas don't warming to the same degree as the more barron areas north and east of the city.. dunno. any info you find on the doctor, I'd be keen to learn more.
Good article.
Thanks guys.
Also regarding sea breezes with no synoptic steering you'll notice that the wind will start blowing straight onshore as the air flows from the ocean to the lower pressure and rising air over land, but as the afternoon and evening progress this will shift anti-clockwise owing to the Coriolis Force.
So on the East Coast an E'ly sea breeze will tend NE after, the southern states, S tending SE and Western Australia, W tending SW.
Ah interesting Craig, I'll keep an eye out for the swing.
Generally speaking, how 'localised' are seabreezes? are we talking something like 15km inland, 50km?
Of the order of 50km rather than 5-15km.
'Storm Force 10' written by Meteorologist John Rousmanierre in about 1976 I think, is a story about the Fastnet yacht race that year, where a 976hp low swept across the fleet in the UK race. From memory, about 1am a 60 knot front went through, the wind veered about 90 degrees, and so you had 30' waves from one direction meeting 30' waves from 90 degrees different. Boats were not just getting rolled, they were getting pitchpoled stern over bow. Besides being a meteorologist, the author was also crew on one of the race boats. There is a fascinating analysis of the storm that he does in the book, unlike anything the met world had ever seen, in which he pieced together the barometric readings and wind records from the 120 or so boats logbook records every 3 hours, it was the first opportunity anyone had ever had of that many barometric readings of a low at sea. The first account I had ever come across that highlit the fact that winds don't simply blow nice and evenly around lows, but that the entire low is a series of troughs and ridges of various strengths, and hence why, of 2 boats only 30 miles apart, one would show double the windspeed of the other, at the same time. He included that in a diagram - far different to what was portrayed by the met bureau.
Fascinating Jody.
Great article and some useful exchanges. Going back to my paragliding days we referred to the morning offshore (cold land, warm sea) as catabatic airflow and the more classical mid day onshore (hot land - cool sea) as anabatic airflow. Meaning in theory that a coastal environment that has some land elevation next to it can be more likely to have an offshore as the cool air falls. So potentially the "fricton" is even more complicated than just "land"? Its also elevation and the variance in Temperature relative to the coastal situation. NZ was classic for this, surfing around Taranaki in winter with a snow capped mountain, we could get offshore in the morning on a coast that was nearly 180 degree semi circle (from Urenui to Hawera).
Great article Craig and the Swellnet Team. Thanks
Differences between the Geelong Racecourse AWS and South Channel Island AWS show the effects of friction very clearly. I always go by SCI.
Today the difference was 10 knots at 10:30.
https://m.
If Swellnet sold a weather e-book that could be updated regularly and also downloadable as a PDF form so it could be printed out I’d buy it for sure.