Jump to content

An alternative climate classification to the Koppen system


Recommended Posts

Rough draft of a possible improved climate classification system...partially adapted from M. Fantamon.

All temperatures reflect the mean (average of diurnal high and low).

When considering a climate classification, choose the best fit if multiple classifications exist.

Temperature conditional, precipitation conditional If bold the condition applies to the entire climate group (ie tropical climates).

 

Equatorial (A)

  • coldest month > 22c
  • annual amplitude <= 5c
  • annual precip >= 2000 mm, wettest month less than 2x wetter than driest month equatorial rainforest (Ar)
  • annual precip >= 1000 mm, wettest month greater than 2x wetter than driest month equatorial gradient (Ag)
  • annual temp >= 25c hot (a) < 25c warm (b*)

Tropical (B*)

  • coldest month > 18c
  • annual amplitude <= 12c
  • annual precip < 800 mm & no month > 80 mm tropical semi-arid (Bq)
  • annual precip < 1250 mm tropical dry (Bd)
  • annual precip < 1500 mm & dry winter/summer, wettest month at least 3x wetter than driest month tropical savanna - dry winter (Bw), dry summer (Bs)
  • annual precip >= 1250 mm, wettest month less than 2x wetter than driest month tropical humid/maritime (Bm)
  • annual precip >= 1500 mm, wettest month >= 400 mm, wettest month at least 4x wetter than driest month tropical monsoon (Bn)
  • warmest month >= 30c desert (a) >= 25c warm (b*)

Subtropical (C*)

  • coldest month between 5c and 18c
  • at least eight (8) months >= 10c
  • warmest month >= 22c
  • annual precip <= 150 mm, no month > 20 mm subtropical arid/desert (Co)
  • annual precip 150-400 mm, no month > 50 mm subtropical semi-arid (Cq)
  • annual precip > 150 mm, total precip < 100 mm over at least three (3) month period, driest month <= 30 mm, wettest month at least 2x wetter than driest month Mediterranean (dry summer) (Cs), subtropical dry winter (Cw)
  • annual precip >= 800 mm, no month < 30 mm, at least one (1) month >= 100 mm subtropical humid (Cr)
  • annual precip >= 1000 mm, wettest month less than 2x wetter than driest month subtropical maritime (Cm)
  • annual temp >= 20c tropical limit / hot desert (a) 15-20c very warm (b*) < 15c warm temperate (c*)

Temperate (D)

  • at least four (4) months >= 10c
  • annual temp >= 3c
  • coldest month <= 0c, annual precip <= 125 mm, no month > 20 mm temperate cold desert (Do)
  • coldest month <= 0c, annual precip 125-400 mm, no month > 50 mm temperate cold semi-arid (Dn)
  • coldest month > 0c, annual precip 125-400 mm, no month > 50 mm temperate semi-arid (Dq)
  • coldest month <= 0c, annual amplitude >= 20c temperate continental (Dc)
  • coldest month <= 5c, annual amplitude >= 15c temperate semi-continental (Dt)
  • annual amplitude 10-15c temperate maritime (Dm)
  • annual amplitude < 10c temperate oceanic (Dr)
  • annual temp >= 10c warm (a), 5-10c mild (b*), < 5c cold (c*)
  • annual precip < 800 mm dry (d), > 800 mm humid (h)
  • wettest month at least 3x wetter than driest month - dry summer (s), dry winter (w)

Subpolar (E)

  • at least one month >= 10c
  • coldest month <= 0c, warmest month < 12c subpolar oceanic (Er)
  • coldest month <= -7c, annual amplitude < 20c subpolar maritime (Em)
  • coldest month <= -30c subpolar cold winter (Eq)
  • annual amplitude >= 20c subpolar continental (Ec)
  • annual temp >= 3c temperate (a), >= -7c cold (b*), < -7c frigid (c*)
  • annual precip < 400 mm dry (d), >= 400 mm humid (h)

Polar (F)

  • no month >= 10c
  • at least three (3) months <= 0c
  • annual temp <= 0c polar ice sheet (Fo)
  • annual amplitude < 20c polar oceanic (Fr)
  • annual amplitude >= 20c polar continental (Fc)

 

Comments and feedback welcome.

  • Like 1

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

Let's examine a case: Portland, OR

Things to consider:

  • Annual mean temp
  • Mean temp of warmest month
  • Mean temp of coldest month
  • Annual precip
  • Wettest and driest month precip
  • Annual thermal amplitude
  • Percent difference between wettest and driest month

For Portland:

 

Annual mean temp: 12.4c
Mean temp warmest month: 20.8c

Mean temp coldest month: 4.7c

Annual precip: 914 mm

Wettest month: 143 mm

Driest month: 17 mm

Annual thermal amplitude: 16.2c
Difference between wettest and driest month: 8.4 (wettest month is 840% wetter than driest month)

Let's figure out the climate zone of Portland based on the proposed classification system:

1. Warmest month is below 22c so Portland is not subtropical under this system. Portland also has only 7 months > 10c (subtropical criteria is 8 months > 10c)

2. Portland meets the temperate requirements (>= 4 months > 10c and annual temp >= 3c)

3. Portland also meets the initial definition for subpolar (at least one month >= 10c) but it fails to meet the requirement that coldest month <= 0c, so it is not subpolar.

 

Portland is temperate (D) as the primary classification

 

Figuring out the secondary classification:

4. Coldest month > 0c so we can rule out Do, Dn, Dc
5. Annual precipitation > 400 mm so we can rule out Dq
6. Annual amplitude > 15c (16.2c) and coldest month < 5c (4.6c) so Portland seems to be temperate semi-continental (Dt) though it is close to being temperate maritime.

 

We can add other specific classifications for temp/precip:

 

Annual temp > 10c (warm): w
Annual precip > 800 mm (humid): h
Wettest month > 3x wetter than driest month with dry season in summer: s

 

Final classification becomes Dt(whs) or temperate semi-continental warm humid dry summer

  • Like 1

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

I like what you've done with the subarctic/subpolar climate. Specifically how you added the -7C isotherm to differentiate between continental subarctic climates and the maritime-influence version. I feel that the subarctic category is too broad under the Koppen system. A location with an average January temperature of -3C should not be classified under the same climate zone as a location that averages -30C (although they may subgrouped by precipitation). There is a Dfd/Dwd subtype that can be found in the colder areas of the Siberian interior, but it doesn't come into play unless the coldest month averages significantly colder (I believe -38C). 

  • Like 1
Link to comment
Share on other sites

Welp, Philadelphia is a subtropical climate and I am living in denial.

Interestingly it's humid subtropical under the Koppen classification as well (the northern extent being roughly NYC on the east coast). Living there for a summer I bet you understand why, haha.

 

Edit: actually the January mean temp is 0.5c in Philly so the proper classification would be temperate (D) rather than subtropical. I'm guessing temperate continental humid Dc(h) as precip is fairly evenly distributed and > 800 mm annually.

 

I got pretty much the expected results for Portland...the climate is very borderline between temperate semi-continental (a continental/maritime hybrid classification) and temperate maritime...the only issue being that there's a very distinct dry summer regime here even at the coast, so it isn't strictly maritime/oceanic really anywhere in Oregon from a precip standpoint.

 

My fear was that Portland could potentially end up with a "Mediterranean" classification which is IMO a subtropical subtype...though Koppen-Geiger includes decidedly temperate locations under the Mediterranean type because precip is the determining factor in their definition, while I hold that temperate dry summer climates are separate from classic Mediterranean.

  • Like 1

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

Up toward Seattle you begin getting to the point where the 10C (50F) threshold becomes an issue also.  Some stations around here have long term averages just below 50 while many have an average in the low 50s.  It can even come down to where some decades average below 50 in areas that nomally average above.

Death To Warm Anomalies!

 

Winter 2023-24 stats

 

Total Snowfall = 1.0"

Day with 1" or more snow depth = 1

Total Hail = 0.0

Total Ice = 0.2

Coldest Low = 13

Lows 32 or below = 45

Highs 32 or below = 3

Lows 20 or below = 3

Highs 40 or below = 9

 

Link to comment
Share on other sites

Up toward Seattle you begin getting to the point where the 10C (50F) threshold becomes an issue also.  Some stations around here have long term averages just below 50 while many have an average in the low 50s.  It can even come down to where some decades average below 50 in areas that nomally average above.

Yes which is why I used the Koppen classification for temperate of at least 4 months >= 50 F but with an additional qualifier of annual mean temp >= 37 F, which puts pretty much everywhere in the US below 1000' elevation in the temperate category...even International Falls, MN which has 5 months on average >= 50 F. Somewhere like West Yellowstone is still subpolar by this definition however a better fit for such a location would be a temperate highland climate variation.

 

in fact Juneau, AK using 1981-2010 normals works out to a borderline temperate climate...it barely ekes out 4 months >= 50 F (June-Sept are 54.6, 56.9, 55.9, 50.0 respectively). In prior 30 year normals Juneau probably fell just barely into subpolar. That shows how far north the west coast remains quite temperate...the entire BC coast is temperate maritime/oceanic for the most part.

 

The subtropical classification seems to work alright. Tulsa would be humid subtropical (Cfa) under Koppen, but is a temperate warm variant under this classification since the coldest month is < 41 F (5c). However Dallas-Fort Worth is subtropical under both classification systems. Even though DFW can see teens occasionally in winter (avg Jan extreme low is 17) the Jan mean is 46 F. Basically my idea of subtropical is the realm of palm trees, including cold-tolerant varieties, that can grow naturally in that region, although on the fringes sometimes that is precluded by frost potential. The goal was to expand the temperate to be more realistic while shrinking the excessively broad subtropical category.

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

I made small adjustments to a few climate sub-types, including adding a couple:

Subpolar oceanic mild winter: warmest month < 12c, coldest month > 0c, annual thermal amplitude < 10c. This type is found in parts of coastal NW Europe (Norway, Scotland) along the immediate coast at or near sea-level. It's also found in a small part of southern Chile around Tierra del Fuego (see Puerto Williams). This distinguishes it from subpolar oceanic, which requires coldest month <= 0c and thermal amplitude < 15c, and separates generally frost-free coastal climates from frost-likely ones. Generally if warmest month > 12c it's more of a maritime or even continental subpolar regime.

Subtropical continental: coldest month < 10c and thermal amplitude >= 18c...areas that fit subtropical criteria but have a bigger annual range and colder winters such as portions of the southern Plains and interior SE US well away from the coast. Fairly small geographic areas that are on the border with temperate zones.

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

So, I guess my climate is humid temperate continental under this system? I like that much better than the humid subtropical crap, haha. I'd argue that our summers are purely tropical/humid tropical, while our winters are humid-continental, if such a description were viable.

 

Cold fronts/continental airmasses rarely (if ever) make it here during July/August, just endless S/SW flow around the Bernuda High, usually. Winter is usually dominated cold, dry W/NW flow out of Canada as the Bermuda High retracts towards the Azores, hence is referred to as the "Azores high" during winter/spring.

  • Like 1
Link to comment
Share on other sites

Yeah that would be my guess

* At least 4 months >= 10c
* Precipitation >= 800 mm
* Annual temp >= 3c

* Annual temp amplitude >= 20c

I added a subtype "temperate boreal" for low elevation locations with coldest month <= -5c which can be used to categorize continental climates in the northern tier of the US and southern Canada (where they also meet temperate requirements).

Yeah I look at the climate of the year as a whole so the hallmark of temperate is having fairly high variations in temp through the year (high degree of seasonality) without a very lengthy warm season (subtropical) or very lengthy cold season (subpolar).

  • Like 1

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

Thanks. Hopefully, some influential individual within the scientific research community stumbles across this and has an epiphany of sorts. ;)

Link to comment
Share on other sites

  • 1 month later...

Juneau would be a "Humid Temperate Subpolar" region, compared to being "a transition zone between a continental climate (Köppen Dfb/Dfc), and an oceanic climate (Köppen Cfb/Cfc), depending on the isotherm used" per Wikipedia.

 

https://en.wikipedia.org/wiki/Juneau,_Alaska

 

Am I getting this right?

"There are so many waves coming in all the time, you don't have to worry about that. Take your time—wave come. Let the other guys go; catch another one." -Duke Kahanamoku

Link to comment
Share on other sites

  • 2 weeks later...

Juneau would be a "Humid Temperate Subpolar" region, compared to being "a transition zone between a continental climate (Köppen Dfb/Dfc), and an oceanic climate (Köppen Cfb/Cfc), depending on the isotherm used" per Wikipedia.

 

https://en.wikipedia.org/wiki/Juneau,_Alaska

 

Am I getting this right?

The airport just barely qualifies as temperate cold humid (4 months at or above 10c). Downtown is a bit milder and is also temperate cold humid. Annual temp is also above 3c which makes it temperate.

 

Along the coast north of Juneau you get into subpolar temperate humid (less than four months >= 10c), such as Yakutat.

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

This is basically a rehash of the Trewartha system with a few tweaks. 

 

I don't like calling the type-C climates "subtropical" if they extend all the way into SW Oregon, which is what this kind of definition allows.  There's nothing remotely tropical about a climate with a midwinter mean temp of 5 or 6 degrees C.  Even if some hardier palm trees happen to grow there ;)

Link to comment
Share on other sites

  • 2 months later...

I think one should start where at least my college physical geography class started (ocean and continent influences), then work from there (moisture, temperature, etc). To that I would add in ecoregion and vegetation. Koppen's system used temperature in some classifications (eastern and tropical climes), then ignored temperature in favor of dryness (his steppe or desert), so that's not helpful. Consistency is helpful.

 

Though it's for horticulture and missed contexts by focusing on California, how about Sunset Wester Garden Book's zones? Or one of the ecoregion systems like Bailey or EPA, with some corrections?

 

To me anything tropical needs to require the lowest record temperature be >32F...maybe the 40's to account for truly tropical plants not being damaged. And divided from equatorial, highland, and those above equatorial. Honolulu and maybe Key West make Tropical, but Miami would be in the warm end of Subtropical?

 

Something that works to embrace most latitudes and continents, with probably distinctives of certain continents giving them their own but related zone...like Perth AU vs. San Diego CA vs. Tel Aviv IS. I can give some sample places later if anyone is interested. What do you think?

 

a. Atmospheric moisture, overall tendencies -

West Coast 

Intermountain (deep continental, surrounded by mountains)

Continental Dry 

Continental Humid (like the midwest to the Appalachains)

East Coast

Upland or Highland

Mountain

 

b. Temperature (w/ finer divisions of the below) -

Tropical

Subtropical

Temperate (or mesothermal to microthermal)

Boreal

Polar / Alpine

 

c. Precipitation (w/ finer divisions of the below) -

Wet

Moist

Semi-arid

Arid

Link to comment
Share on other sites

  • 2 weeks later...

This is basically a rehash of the Trewartha system with a few tweaks. 

 

I don't like calling the type-C climates "subtropical" if they extend all the way into SW Oregon, which is what this kind of definition allows.  There's nothing remotely tropical about a climate with a midwinter mean temp of 5 or 6 degrees C.  Even if some hardier palm trees happen to grow there ;)

Show me where in Oregon has 8 months with a mean temp >= 10c AND warmest month mean temp >=22c.

 

Brookings has 9 months >= 10c but no month >= 22c.

Medford has 2 months >= 22c but only 7 months >= 10c.

 

Ukiah, CA on the other hand meets both requirements.

The Pacific Northwest: Where storms go to die.

Link to comment
Share on other sites

  • 8 months later...

I think it should be based entirely on vegetation (and equating vegetation with similar growing requirements). There are just too many factors to classify climates with measured variables. In addition to temperature, precipitation, and the like is latitude. For example, at Clinton, BC (51N) you have ponderosa pine growing all over the place even at above 3000 ft, but you go up to Williams Lake (52N) where it's warmer and drier and more conducive to ponderosa on paper, you don't get ponderosa because it's too far north. 

Link to comment
Share on other sites

  • 1 year later...

I am looking for a narrative daily, weekly or monthly summary of weather conditions that have occurred like this.

 

 

18-24: Weekly temperatures averaged at least 5°F above normal from Montana to northwestern Kansas.  Rainfall totaled 2 inches or more in portions of the Mid-Atlantic States.  In addition, cold conditions dominated the eastern half of the country for much of the week, with bitterly cold air sweeping across the Great Lakes and Northeastern States on Thanksgiving Day.  Weekly temperatures averaged 10 to 15°F below normal across much of the Northeast.

Link to comment
Share on other sites

  • 1 year later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...