Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
you are wrong, -0.5°C rounds to -1°C and 17.5°C rounds to 18°C. what makes 17.5°C so special that it gets to round to the threshold while -0.5°C doesn't? they are both equally far away from their corresponding thresholds...
That convention might be used by some people but it is very inconsistent, and I always consider rounding 'x+0.5' to the higher value 'x+1', regardless of whether 'x' is positive or negative. So, I would still consider -0.5 to round to 0.
In any case I don't care about those contrived conventions. This discussion digresses from my original point, which was simply that using 0.1C precision is not always sensible as most temp gauges are not tuned to that kind of accuracy. 1C scaling more sense to me.
That convention might be used by some people but it is very inconsistent, and I always consider rounding 'x.5' to the higher value x+1, regardless of whether x is positive or negative. So, I would still consider -0.5 to round to 0.
In any case I don't care about those contrived conventions; who came up with them anyways? This discussion wavers from my original point, which was simply that using 0.1C precision is not always sensible as most temp gauges are not tuned to that kind of accuracy. 1C scaling more sense to me.
fair enough. i prefer your method anyways but i don't use it. my rule for rounding is four decimal places, and i don't really care if i go up or down, because its four decimal places and literally no one cares at that point unless you are calculating a planet's trajectory for a space mission or something way more important than this...
ultimately i feel 0.9°F (or in Fraser's case 0.54°F) is still quantifiable... and in Fahrenheit 63.86°F doesn't even round to 64.4°F...
Last edited by Sir Goosenseresworthie; 12-31-2015 at 12:16 AM..
ultimately i feel 0.9°F (or in Fraser's case 0.54°F) is still quantifiable...
OK, fair enough.
Quote:
Originally Posted by Sir Goosenseresworthie
and in Fahrenheit 63.86°F doesn't even round to 64.4°F...
That's a non-issue. If we use 1C significance for our purposes then we wouldn't convert to F in the first place. Why would you change units in the middle of a computation?
If you want to use Fahrenheit to begin with, and if you use 1F significance, both values would become 64F. So, still tropical.
Honolulu is a superb climate among the overall tropical area in the earth, the Hawaiian Islands in general rarely receives convective rain a.k.a. severe thunderstorms //www.city-data.com/forum/weath...derstorms.html and also constantly maintains mild to warm temperature range for the latitude which the city don't see below high-50s°F(14~15°C) or above low-90s°F(32~33°F) most time of the year.
That's a non-issue. If we use 1C significance for our purposes then we wouldn't convert to F in the first place. Why would you change units in the middle of a computation?
If you want to use Fahrenheit to begin with, and if you use 1F significance, both values would become 64F. So, still tropical.
but where does it say to use a 1° significance at all? it seemed like the only reasoning for counting Fraser as tropical was the fact that it was within some arbitrary rounding distance. i like to think the threshold was created for the purposes of requiring a definite threshold, and if they wanted places with temperatures 1° lower than the threshold to count, they would have lowered the threshold by 1°
If you want to use Fahrenheit, you also round the Celsius translated equivalent threshold of 64.4 to the nearest whole number in Fahrenheit; 64F. 63.86 is closer to 64F than 17.7 is to 18C; therefore Fraser Island is tropical even moreso in Fahrenheit than in Celsius. I'll admit that it's on the cooler side of tropical but definitely meets the standards for tropical and this thread. I'll do separate subtropical climate island battle sometime in the future.
Last edited by ABrandNewWorld; 01-05-2016 at 09:21 PM..
If you want to use Fahrenheit, you also round the Celsius translated equivalent threshold of 64.4 to the nearest whole number in Fahrenheit; 64F. 63.86 is closer to 64F than 17.7 is to 18C; therefore Fraser Island is tropical even moreso in Fahrenheit than in Celsius.
what, i swear you people make up your own rules. now you can round down numbers?
Quote:
Originally Posted by ABrandNewWorld
I'll admit that it's on the cooler side of tropical but definitely meets the standards for tropical and this thread. I'll do separate subtropical climate island battle sometime in the future.
no it doesn't, and when you make the subtropical climate island battle Fraser should be there, because it actually falls under the subtropical classification (without any data manipulation) and does not fall under the tropical classification without rounding. therefore it is a better fit for that battle than it is for this one.
but where does it say to use a 1° significance at all? it seemed like the only reasoning for counting Fraser as tropical was the fact that it was within some arbitrary rounding distance.
Nothing arbitrary about it. When trying to determine a climate's Koppen classification, the correct system of units to use is Celsuis, not Fahrenheit, and the correct significance is 1, not 0.1. This is implicit in the definition of the Koppen classification system (for example, like G8R said, the threshold for tropicality is 18C, not 18.0C. Nor is it 64F).
So, my example with Fahrenheit was hypothetical; if I'm working with Koppen I wouldn't convert to Fahrenheit in the first place.
In fact it actually depends on the source. In this source it says you can use both, with 1 degree significance on both scales, not 0.1: http://www.eoearth.org/view/article/162263/
Quote:
Originally Posted by Sir Goosenseresworthie
what, i swear you people make up your own rules. now you can round down numbers?
What? 64.4 definitely rounds to 64. Nothing new about that.
But like I said, for Koppen I personally wouldn't use Fahrenheit in the first place, as it could lead to inconsistency.
Nothing arbitrary about it. When trying to determine a climate's Koppen classification, the correct system of units to use is Celsuis, not Fahrenheit, and the correct significance is 1, not 0.1. This is implicit in the definition of the Koppen classification system (for example, like G8R said, the threshold for tropicality is 18C, not 18.0C. Nor is it 64F).
So, my example with Fahrenheit was hypothetical; if I'm working with Koppen I wouldn't convert to Fahrenheit in the first place.
In fact it actually depends on the source. In this source it says you can use both, with 1 degree significance on both scales, not 0.1: Köppen Climate Classification System
What? 64.4 definitely rounds to 64. Nothing new about that.
But like I said, for Koppen I personally wouldn't use Fahrenheit in the first place, as it could lead to inconsistency.
Would Fraser Island be tropical without data manipulation? Nope. I don't care how close 17.7 is to 18.0, 17.7 is STILL smaller than 18.0 therefore Fraser Island isn't tropical. And I am questioning eoearth's view on using a 1°C significance. I do agree that °C should be used with Koppen.
And I am questioning eoearth's view on using a 1°C significance.
I feel like further discussion about this would be pointless, without knowing what kind of significance is actually used in practice by climatologists and the like. Personally, I feel like 0.1 wouldn't be correct because the error bars would be wider than that. But that is disputable and I acknowledge that some might people might use 0.1 significance. I wonder if someone could find a source on it.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
