Koppen-Grega94 Climate Classification System (snow, hot, warmest, ice)
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.
This is my attempt at improving Koppen's climate classification system, but my attempt isn't to overhaul it and add tons of more categories, but just to add a few tweaks, otherwise it's basically the same classification system. if my definitions end up being very similar to something someone already has done, then feel free to ask the moderators to combine the threads, otherwise here it goes...
So my biggest issue with Koppen is the B arid climates and the C temperate climates, the other groups I will be leaving alone for the most part. My issue with the arid climates is that they are very broad and can easily be classified with the second variable that denotes precipitation. And for C climates as many others have pointed out is too broad. Now I'm going to go back to Koppen's original definition of -3C and use that as the base since that is the threshold when permanent winter snow pack typically starts (and when you know that Koppen lived in St. Petersburg it makes a lot of sense why he chose that isotherm). Now comes the issue on how to break this C group. Ideally I would want to split it into two categories, the temperate one that has a dormant winter season with no permanent snow pack but still experiences snow regularly, and the subtropical group that experiences almost no snow and vegetation can grow in the winter (or perhaps an extremely short dormancy period?). I'm not sure exactly where would be the best place to split it, but for now I will split it in half at 7.5C. But it could be based off of the temperature that plants usually grow which I think is 10C? Using 10C as the isotherm makes a lot of sense, since it's used to define the polar climates, and would do away with the cold "c" climates in the new B subtropical group, which would be a bit of an oxymoron. please vote in the poll on which isotherm(s) to use. At the very bottom are some borderline examples that can help you decide on the isotherm.
So now that I explained my reasonings here are the variables
A - Tropical: > 18C in the coldest month
Af - Tropical rainforest
Am - Tropical monsoon
Aw - Tropical savanna (dry winter)
As - Tropical savanna (dry summer)
Ad - Tropical dessert (arid) *tropical version of BWh (w is already taken and got d from dessert)
Ag - Tropical semi-arid/steppe * tropical version of BSh (s is already taken and got g from grassland) (not sure if this is necessary as savanna climates are already somewhat semi-arid)
B - Subtropical: 7.5C (or 10C) to 18C in the coldest month
------------------------
Bfa - subtropical, humid, hot summer
Bfb - subtropical, humid, mild summer
Bfc - subtropical, humid, cold summer *doesn't exist if 10C isotherm is used
-------------------------
Bwa - subtropical, dry winter, hot summer
Bwb - subtropical, dry winter, mild summer
Bwc - subtropical, dry winter, cold summer
-------------------------
Bsa - subtropical, dry summer, hot summer
Bsb - subtropical, dry summer, mild summer
Bsc - subtropical, dry summer, cold summer *doesn't exist if 10C isotherm is used
-------------------------
Bda - subtropical, desert, hot summer
Bdb - subtropical, desert, mild summer
Bdc - subtropical, desert, cold summer *doesn't exist if 10C isotherm is used
-------------------------
Bga - subtropical, semi-arid, hot summer
Bgb - subtropical, semi-arid, mild summer
Bgc - subtropical, semi-arid, cold summer *doesn't exist if 10C isotherm is used
C - Temperate: -3C to 7.5C (or 10C) in the coldest month
*use the above variables, just replace the subtropical with temperate
D - Continental: < -3C in the coldest month
*use the above variables, just replace the subtropical with continental
E - Polar < 10C in the warmest month
ET - Tundra
EF - Ice cap
Ed - Polar desert *Polar version of BWk (not sure if this is necessary)
Eg - Polar semi-arid *Polar version of BSk (not sure if this is necessary)
Here are some of the borderline climates depending on if we use the 7.5C or 10C, and for good measure will also add ones that fall between -3C and 0C, as I'm sure some will disagree on that limit as well.
Don't agree with the Scilly Isles being subtropical, while my climate is temperate - there will never be able to grow the range of subtropical species here does, and snow is rare here - 3 times in the last 120 years.
That is to be expected around the margins though, and overall I think this better identifies broadly similar climates, quite well.
Don't agree with the Scilly Isles being subtropical, while my climate is temperate - there will never be able to grow the range of subtropical species here does, and snow is rare here - 3 times in the last 120 years.
That is to be expected around the margins though, and overall I think this better identifies broadly similar climates, quite well.
I did forget to mention in the polls if the isotherm should be below 7.5C, do you think it should be decreased, perhaps down to 5C?
I agree that subtropical needs to be split, and the isotherm should be close to 9 C / 48 F (evenly splitting 0 C / 32 F and 18 C / 64 F), so I picked 10 C / 50 F. As for the boundary between humid continental and temperate, I'm thinking 0 C / 32 F because it's a natural dividing line, being the freezing point of water. The "persistent snow line", which was the logic for -3 C / 27 F, is going to vary wildly with climate.
I wholeheartedly agree with taking the "B" category and making it a subtropical category, and putting arid climates under their corresponding temperature bands. That's the biggest problem with the current Koppen system - A, C, D and E are temperature-based classifications while B is precipitation-based. This system makes it consistent.
I agree that subtropical needs to be split, and the isotherm should be close to 9 C / 48 F (evenly splitting 0 C / 32 F and 18 C / 64 F), so I picked 10 C / 50 F. As for the boundary between humid continental and temperate, I'm thinking 0 C / 32 F because it's a natural dividing line, being the freezing point of water. The "persistent snow line", which was the logic for -3 C / 27 F, is going to vary wildly with climate.
I wholeheartedly agree with taking the "B" category and making it a subtropical category, and putting arid climates under their corresponding temperature bands. That's the biggest problem with the current Koppen system - A, C, D and E are temperature-based classifications while B is precipitation-based. This system makes it consistent.
Now that you mention it that limit does seem to be quite a bit finicky and would need to be recalculated to be a function of temperature and precipitation since there are places that definitely have permanent snow cover, but also are above -3C such as Paradise, WA which is only -2.6C in it's coldest month, but gets well over 17 meters of snow a year, and no amount of daytime warmth will manage to melt all of that in the middle of winter, so it's probably best to avoid that and set it at 0C.
Now that we set that out, how about 9C vs 10C? using 9C is pretty simple as it's the halfway point, but one degree warmer and you can avoid the Bxc group all together and reduce the amount of categories, but then again how many would you find that fits that criteria? Is there a noticeable difference between climates that are above 10C all year round and those that aren't?
Is there a noticeable difference on winter vegetation between Charleston, SC and Jacksonville, FL?
I think this classification is definitely an improvement on Kpppen and I like it. The boundary between the B and C climates should be 10C as that is the isotherm for tree growth in polar climates and would reflect a nearly year round growing season while still having a noticeably cooler period of the year. Between C and D, it should be 0C as the freezing point of water seems like a good natural boundary and would show where frozen precipitation is more dominant in the winter compared to liquid precipitation.
Another thought that came to mind is moving the tropical threshold further up. Currently it’s at 18C, but if we are going to use 10C for the lower limit of the B category due to it already in use to define polar influences, then perhaps for the sake of constancy it should be 22C ,which is used to define hot summers which I guess implies tropical influence, wouldn’t it make sense to use the same isotherm for both tropical climates and hot summers? What is the reasoning behind the two? Perhaps hot summers should be lowered to 18C? or maybe meet half way at 20C and that way the B and C groups have an equal temperature range.
Set the winter threshold too high between B and C high for a winter limit, and it's back to the situation of saying a climate that produces and avocados and passionfruit with ease, and can have jacaranda lined streets and 12 months of grass growth, is no different to the likes of Copenhagen, Bergen or Amsterdam.
Trewartha's 8 months>10C is superior for identifying higher levels of winter ecological activity.
Set the winter threshold too high between B and C high for a winter limit, and it's back to the situation of saying a climate that produces and avocados and passionfruit with ease, and can have jacaranda lined streets and 12 months of grass growth, is no different to the likes of Copenhagen, Bergen or Amsterdam.
Trewartha's 8 months>10C is superior for identifying higher levels of winter ecological activity.
So then should the B and C groups have the same temperature range (0C to 18C) but the only difference is that the B group has 8 month >10C and the C group has between 7 and 4 months > 10C?
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.
