psyche_da_mike24's Climate Classification System (hottest, warmest, ice)

[psyche_da_mike24]

I was inspired by QIDb602's thread to make one on my own climate classification system that I created a while back.

TL;DR- basically a mashup of Köppen and Trewartha with a more nuanced precipitation threshold.

The main distinguishing feature or innovation of my system is that the semi-arid and arid climates are automatically assigned to a primary temperature regime (tropical, subtropical, etc.) The second major innovation is temperature-based subcategories for Tropical climates to better distinguish altitude levels and heat regimes.

I think it's important to create a climate classification system centered around human comfort and survivability, especially in a world that is facing rapid and unpredictable climate change in the foreseeable future.


First Letter- primary temperature letter (focused on coldest month). I used 10 C as a threshold for subtropical climates because it's already used to define growth seasons in subpolar and polar climates.

A = Tropical: all months > 18 C

B = Subtropical: all months > 10 C, coldest month < 18 C

C = Temperate: 4-11 months > 10 C, coldest month -3 C to 10 C

D = Continental: > 3 months > 10 C, coldest month < -3 C

E = Subarctic/Subalpine: <= 3 months > 10 C

O = Polar/Alpine: all months < 10 C

OT = Polar Tundra (at least 1 month > 0 C)
OF = Polar Ice Cap (all months < 0 C)


Second Letter- precipitation letter. The dry-summer and dry-winter thresholds are the same as Köppen's. I got the idea to change the tropical rainforest climate threshold from Bisfbath's Af/BSh thread.

For dry climates:
- R = arid (Gk. xeRos): actual precip. less than 0.5 of precip. threshold
- Z = semi-arid (Gk. aZaleos): actual precip. between 0.5-1.0 of precip. threshold
For tropical:
- f = rainforest: different from Köppen (<= 3 months <100 mm, instead of all months >60 mm)
- m = monsoon: same as Köppen (driest month greater than (100-(total/25)) mm)
- w = savanna: same as Köppen (driest month less than (100-(total/25)) mm)
- s = same as above but with summer dry season
For non-tropical:
- s = dry summer: (wettest month > 3 * driest month, driest month < 40 mm)
- w = dry winter: (wettest month > 10 * driest month)
- f = default: no other conditions are fulfilled


Third Letter- secondary temperature (focused on warmest month). I added a third letter for Tropical climates to distinguish between equatorial climates (which have virtually no seasonal temperature variation) and non-equatorial tropical climates, and to distinguish very hot tropical climates (>28 C) like those found in the Asian continent.

For tropical:
- A = very hot: coldest month > 22.5 C AND annual average > 28 C
- e = equatorial: average monthly temp range < 3 C
- E = very hot equatorial: equatorial if annual average > 28 C
- b = cool: annual average < 22.5 C [optional]
- a = default: does not fit any of the other categories

For non-tropical:
- a = hot-summer: hottest month > 22.5 C
- b = warm-summer: hottest month 10 C - 22.5 C

For subpolar/subalpine only:
- b = equatorial: coldest month > 6 C [optional]
- c = maritime: coldest month -3 C to 6 C
- d = continental: coldest month < -3 C

For polar/alpine only [optional]:
- m = mild: coldest month > -3 C


Precipitation threshold: adjusted for average annual temperature and seasonality. The threshold increases faster at higher temperatures due to higher potential evapotranspiration and increased risk of desertification.

Let threshold Q be a function of T and U where T = average annual temperature and U = (sun-high season precip)/(annual precip).

if (T < 10 C) {Q = 100 + 10*T}
else if (T < 20 C) {Q = 200 + 20*(T-10)}
else if (T < 30 C) {Q = 300 + 30*(T-20)}
else {Q = 400 + 40*(T-30)}

if (U < 0.1) {Q += 0}
else if (U < 0.9) {Q += 350*(U-0.1)}
else {Q += 280}

If the threshold Q < 0 then the climate is non-dry by default.


Optional "extreme" designation:

For tropical: >= 6 months average temp > 32 C, must fulfill conditions for third letter A or E
Created with the Red Sea and Persian Gulf regions in mind. Could feasibly appear in South Asia under extreme warming scenarios

For subtropical: >= 3 months average temp > 32 C
Created with the Greater Sahara region and West + South Asia in mind. Could feasibly appear in China under extreme warming scenarios

For subpolar/subalpine: coldest month < -35 C
Created with the Russian Far East in mind, hypothetical category for hyper-continental high-latitude climates

For polar/alpine: coldest month < -50 C, >= 3 months < -40 C
Created with Antarctica in mind, hypothetical category for continental ice sheets



Examples:

London: Cfb
New York City: Cfa
Beijing: Dwa
Dubai: ARa/ARA
Hong Kong: Bwa
Paris: Cfb
Shanghai: Cfa
Singapore: Afe/AfE
Sydney: Bfa
Tokyo: Cfa
Bangkok: AmA/AmE
Istanbul: Csa
Los Angeles: Bsa/BZa
Mexico City: Bwb
Moscow: Dfb
Madrid: Csa
Miami: Awa
Johannesburg: Cwb
Lisbon: Bsa
Montreal: Dfb/Dfa
Riyadh: [xtrm] BRa
Santiago: CZb/Csb

Spoiler

BONUS-
Lima: BRb
Vostok: [xtrm] OF
Yakutsk: [xtrm] Efd
Dallol: [xtrm] ARA
Hyderabad India: Awa/AZa
Hyderabad Pakistan: [xtrm] BRa/ARa

[Joe90]

Not seeing commonality in one C climate that has 4 months >10C, and winters averaging -2.5C, compared to another C climate that has 11 months>10C and winters averaging 9.5C -these climates would be worlds apart in human comfort, and the ability to survive.

Temperature thresholds only really make sense in the context of the movements of air masses

[psyche_da_mike24]

Originally Posted by Joe90

Not seeing commonality in one C climate that has 4 months >10C, and winters averaging -2.5C, compared to another C climate that has 11 months>10C and winters averaging 9.5C -these climates would be worlds apart in human comfort, and the ability to survive.

Temperature thresholds only really make sense in the context of the movements of air masses

Winter =/= coldest month, and my temperate category is really a catch-all for everything that isn't definitively subtropical, subpolar, or continental.

[Joe90]

Originally Posted by psyche_da_mike24

Winter =/= coldest month, and my temperate category is really a catch-all for everything that isn't definitively subtropical, subpolar, or continental.

A climate with 11 months>10C and winters at 9.5C, would likely have much influence from subtropical air masses, and vegetation that would be more similar to vegetation in subtropical regions, while 4 months>10C and -2.5C winters, would likely have predominantly polar weather with very different vegetation.

Probably better to try and have classifications that share a commonality, movement of air masses or vegetation etc.

[psyche_da_mike24]

You have to draw the line somewhere. I think the 8 months > 10 C threshold for subtropical is a good poleward boundary for defining subtropical climates, but would argue that a hard 10 C threshold for the coldest month is better for delineating climates that are actually sub-tropical (i.e. resemble the tropics in their lack of winter cold).

[Joe90]

Originally Posted by psyche_da_mike24

You have to draw the line somewhere. I think the 8 months > 10 C threshold for subtropical is a good poleward boundary for defining subtropical climates, but would argue that a hard 10 C threshold for the coldest month is better for delineating climates that are actually sub-tropical (i.e. resemble the tropics in their lack of winter cold).

The C category doesn't delineate winter cold though - Nice and Oslo can't be said to have similar winters or a similar experience of cold.

An 11C winter can still have plenty of cold - many climates with winters that warm, can't grow the same vegetation as my colder area, because it's actually my area that lacks the winter cold.

[Neptunepenguins]

A 10C winter might have a day where it is -5C at 3 PM and a low temperature of -10C that same night. Furthermore, that same place has a record low of -14C. Now that isn't to say that maybe in the warmest winter month there will be days of 32C like in February of 2017

[Bisfbath]

I like it, it's an improvement on Köppen's system in my opinion, especially on his Cf* categories which to me are the climate equivalents of wastepaper basket taxons. They are too broad, which is especially noticeable as his Af category is too narrow, as we've seen in the rainforest having at least 9 months of >100 mm rainfall rather than all months >60 mm.

[Mhc1985]

Sorry for being the bad guy here.

I think it's a mess. We end up with too many categories.
I counted a maximum of 82 climate types, and going up to 97 if we consider the optional ones (obviously there must be quite a bit which don't occur naturally).

The idea of making up a classification is grouping the units in various sets, not making up a set for every unit.
A more interesting exercise would be putting a limit in the amount of climate types before designing the system, so we have to carefully think what we could do to make it detailed enough without looking excessively broad or lazy. For instance, try to create a climate system with only 10 types. Then enhance it to 15 or 20. Köppen has 30 types, why not rearranging these 30 into a less disperse grouping?

[Osmium]

Yeah, I'd have to agree there are probably too many subcategories in this scheme. There comes a point where there are so many options that a classification scheme becomes somewhat pointless.

While these sorts of schemes (Koppen, Trewartha) will never nail down all locations perfectly, they do a decent summary of how region affects climates around the world. You may as well be looking at a micro-climate diagram or map then with so many options.