mountain weather forecasting

Cool, thanks!

Hi davidm. I just found the Mitre (Tararuas) forecast page at http://www.mountain-forecast.com/peaks/Mount-Hector-2/forecasts/1529 I'm not sure where it's pulling the forecast and the various charts from, and I'm no expert on forecasting, but I'd be cautious about trusting it without knowing. It looks as if they've fed a database of mountain locations into the website, and it may be simply auto-generating forecasts direct from the global weather data that comes from one of the 2 big model runs out of the UK or the US, without the same kind of fine-tuning for NZ that gets applied to the auto-generated maps available through sites like MetVuw.com and the automated stuff at MetService.co.nz. The front page does say they're running weather algorithms, but if they've not tuned them pretty well for NZ's specific geography, the forecasts are likely to be worthless as soon as a situation starts getting slightly complex. The Met Service often mentions the Tararuas in their Brief and Extended NZ mountain forecasts which are issued for free on its website, and the Tararuas in particular started getting more attention from a year or two ago after they ran a survey and lots of people asked for it. Have you seen this? For some reason it's necessary to go into one of the National Park forecasts, then read the NZ forecast in the pane down the right-hand side. The forecast there right now even says "Light winds, but strong westerlies about the Tararua tops." The global data from the US and UK basically treats NZ as a tiny blob represented by about 4 points of land on the global map that average something like 300 metres high. It's impractical to expect any forecasts obtained directly from those kinds of assumptions, which is why it's necessary to run lots of extra NZ-specific modeling over the top of the global data before sites like MetVuw or MetService publish it. The front page of Mountain-Forecast says it's running algorithms that are "proven to work" for thousands of ski resorts and surf breaks, which implies the mountain part of those algorithms was designed for areas with large land masses (Europe/USA) which happen to have lots of ski resorts.

Hi there izogi Thanks for the post. Yes, I'm an avid reader of the metservice mountain forecasts and it's true that they do mention the Tararuas as part of their North Island summary, but it's not nearly detailed enough and there are many days when the Tararuas get lumped in with every other range when as we locals know only too well, they're a beast unto themselves. I'm just back from a trip to Aokaparangi and Mid Waiohine and the weather changed about every five minutes. You may well be right about the US website, but it's interesting that they are at able to give such detailed information for e.g. Mount Hector and Mitre when metservice apparently can't. I'm sure there are plenty of people like me who'd happily pay a reasonable annual subscription to get the kind of information that's available for free on the US website overlaid with NZ-specific data. What's true for the Tararuas must also be true for every other mountain range. Maybe someone should talk to metservice about a premium service for trampers.

I have observed that Metservice forecasts have generally been very accurate in recent years. Of course there are the occasional clangers but I rememeber one weekend recently the weather changed as predicted almost by the minute as the weekend progressed, and that was on Tongariro which has been notoriously difficult to predict at times. One thing that even the Metservice cannot predict is microclimates. Often the weather in a particular location can be effected by local geography in funny ways. You need local knowledge to understand those. A general understanding of how NZ weather works is useful in the hills when a forecast isn't immediately available. For example inthe3 Southern Alps when it is raining and there is a sudden drop in temparature and even snow, yhou can start smiling because it is a SW change and the weather will clear. One advance I like is the weather radar. I know it isn't a forecast but I think it is awesome seeing the weather moving up the country. Maybe one day those images will be available on a website suitable for viewing on a cell phone. That would be great for trampers in cell phone range anyway.

It'd be interesting to follow and maybe try to do some objective tests. It's very tricky to design useful tests given the nature of how weather in NZ can be radically different within quite short distances (as pmcke pointed out with microclimates). For the specific peaks, I'm guessing they'd just be plugging a latitude, longitude and elevation into whatever northern-hemisphere-large-land-mass-optimised model they have, and getting whatever (possibly useless) data it pumps out, which might even be assuming that New Zealand doesn't have a mountain higher than a few hundred metres. I might be very wrong, but they're also very unclear about exactly what they're doing and the global nature of the site suggests it doesn't pay much attention to NZ. On a tangent, something to keep in mind with auto-generated forecast charts (such as the forecast charts on MetVuw and the MetService for wind and precipitation) is that they're often quite good, but there are certain kinds of weather situations that the algorithms simply can't predict reliably. If you watch the charts over several days, leading up to the day you care about, and the charts appear stable, it's likely that they're fairly accurate. If the charts' predictions fluctuate and change lots during the time leading up to that day, it's likely to be one of those situations where the predictions are unreliable, so whatever's being generated, even the day before, could be out by quite a lot.

"For example in the Southern Alps when it is raining and there is a sudden drop in temperature and even snow, you can start smiling because it is a SW change and the weather will clear" Useful info - where can I get more like this?

That mountain forecast dot com site is run by a New Zealander based in London. He does the excellent snowforecast.com site. We've been using it for years and find it is often neck to neck in predictions with the Metservice site. The snow forecast dot com site is great for wind direction, strength, precipitation - rain versus snow, cloud cover and temperatures. Frank thinks they probably don't take orographic effect into account i.e. eastern rainshadows. A good way to develop an understanding of the weather is to keep a record of predictions and outcomes for a year if you're pedantically inclined. Not that I have (unlike Frank) but I read Eric Brenstrum's NZ Wx book a couple of times and that helped. Bob McDavitt runs good weather education articles in the FMC Bulletin which can be got from a public library. The November 2010 article talks about the orographic effect on winds from the west in the Southern Alps and the narrower lower Tararuas.

Thanks Honora. I couldn't find any info on the site about who actually runs it or what the methods are, but it's nice to know there's a person involved who probably recognises New Zealand actually exists as a land-mass. :) I tried to read Erik Brenstrum's book a few years ago, but had a lot of trouble with the constant frequent references to and isobar maps of obscure weather events in the past several decades, and eventually gave up. (I know they're all to demonstrate all sorts of things, but I had lots of trouble following it at the time.) I should probably try it again some time, I think it's somewhere on the shelf behind me. There's also a MetService blog which occasionally has interesting posts, at http://blog.metservice.com/

Cheers, izogi. Just subscribed to it.