Temperature & Climate
Recently, I read an article about temperature and climate charts for D&D. This is one of those times when you have to ask yourself if the benefit outweighs the effort. On a day to day basis, I really don’t see the usefulness of detailed climate information. But as a generality, I think it is important to at least have a rough idea of what to expect in a region.
There is nothing wrong with static weather. So what if every day is sunny and clear, with moderately warm days, comfortably cool nights, and amazingly, all the plants are green and thriving despite the lack of rain. It’s a fantasy game where you are busy fighting monsters and saving fair damsels. Is it really necessary to worry about the weather?
On a day to day basis, I would say no. But you should be familiar enough with climatology to know what is reasonable weather and what isn’t. In northern Alaska, you wouldn’t see a month long 120° heat wave. In an equatorial rain forest, you wouldn’t see a blizzard. Recognizing such extreme anomalies may seem like common sense, but where is the cutoff where heat waves/blizzards are reasonable?
My Old Charts
I have some very detailed charts, based on those that Gary included in his World of Greyhawk folio, that predict: temperature, humidity, and wind speed. They take into account terrain type, elevation, latitude, and wind chill. Although not perfectly accurate, they offer a very workable model. After a great deal of work fine-tuning them, they got stuck in a folder and never used. I just never needed that level of detail and the amount of work required to use them was prohibitive.
An Overview
A quick search for climate shows up a wiki page on Climate Charts. A climate chart shows the average range of temperatures and precipitation, by month, for a given region. Here are sample charts of three regions, as well as a description, taken from that site:
As we can see from the chart, Maribor has a temperate climate with hot summers and freezing winters. It lies in the northern hemisphere, so the temperatures peak in July and August. The temperature in Labuan, which lies in the heart of the tropics, hardly changes through the year. Instead of summers and winters, there is a dry season in the beginning of the year, followed by a wet season with high rainfall. Cuzco also lies near the equator, but at a much higher altitude in the Andean highlands, and also much drier. Like in Labuan, the daily high temperature barely changes through the year, but they are significantly cooler due to the altitude. Nights in Cuzco are much colder than during the day, especially in the dry months from May to August.
Most of the US more or less resembles the first chart. The range of temperatures remains fairly constant each month, while the average temperature raises significantly in the summer and drops in the winter. This climate type is all I am familiar with so, subconsciously, I assume that it is representative of a global condition. As you can see from the other charts though, this isn’t true. Near the equator, where the angle to the sun never becomes extreme, there is little variation between summer and winter. A little research will show that there are actually a number of factors that make it difficult to accurately describe planet-wide climate in a single model.
A Compromise
The single most useful climatological piece of information we could generate would be the temperature. From that, we can extrapolate whatever else we need. In fact, I wouldn’t suggest actually using such charts to generate temperature so much as use them for a guide as to what ranges of temperatures are reasonable and then assign them, on the fly, as befits the situation.
| Month | Average Temp |
|---|---|
| Jan | 33° |
| Feb | 43° |
| Mar | 55° |
| Apr | 68° |
| May | 77° |
| Jun | 80° |
| Jul | 77° |
| Aug | 68° |
| Sep | 55° |
| Oct | 43° |
| Nov | 33° |
| Dec | 30° |
Okay, enough talk. Let’s get started. I want to base my model at 45° latitude with an average daily high of 80°F in the summer and 30°F in the winter. I’m going to use a sine curve to vary the temperature between those two extremes, which gives the Average Temp on the chart to the left. From this average temperature, I will subtract 10°F to create a Base Temp. Then, to find the high and low for the day you will add 1d20 for the high or subtract 1d10 for the low. This will generate the average daily highs I desire, with an average daily low about 15°F cooler.
There is no method of accounting for different latitudes that is both easy and accurate. So instead, I’m going with what’s easy. Between 30° and 60° latitude, 1° of change in latitude equates to 1°F of change in temperature (cooler as you move toward the pole, warmer as you move toward the equator). Each degree of latitude above 60° (toward the pole) drops the temperature by 2°F, while every 2° of latitude below 30° (toward the equator) raises the temperature by 1°F.
For being such a crude method, this technique isn’t too bad. It generates average daily temperatures between -45° (winter) and 5° (summer) at the pole, and between 60° (winter) and 110° (summer) at the equator. In real life, the north pole ranges from -15° to -45° in the winter, and stays around 0° in the summer, while the equator maintains a temperature of roughly 90° all year round.
Using This Info
With the chart above, you can simply look up the month and see what the average daily high should be. The average daily low will be about 15°F cooler. With that information, I’d suggest simply “winging it” whenever the subject comes up. As I said earlier on, I don’t see much use in worrying about the weather. However, it can be fun to mix things up once in a while. Throw in a snow day now and then. From the chart above, you can see that snow is reasonable all winter long and in late fall or early spring. That’s pretty much common sense for the latitude given. But if you change the latitude significantly, this will provide guidelines for how far that snow zone can reasonably be shifted.
As far as actual climate (blizzards, fog, wind storms, etc), I think those sorts of things are best placed as needed, instead of being left to the luck of the dice. When travelling high in the mountains, constant snow (occasionally turning to blizzard) makes a good story. Perpetual rain when travelling through murky swamps or heavy fog on the moors just adds atmosphere.
Keep in mind that blizzards, heavy rain, and heavy fog significantly reduce visibility. Finding that temple in the marsh may seem easy when looking at a map in town, but sloshing through the mud in a heavy rain, with visibility dropping down to almost nil, is an entirely different story.
This is an area of fascination for me, so bear with me if I add a bit of complication to your idea.
As you have it here, there is a 50º F variation in the average daily high between summer and winter. Perhaps with each degree you move toward the equator, not only does your mean daily high increase by 1º, but your variation decreases by one degree. Vice versa as you move away from the equator.
Thus, at 45º latitude, temperature varies from 30ºF to 80ºF, as you have it. At the north pole the mean daily high temperature remains 5ºF at its peak, but the dives to -90º F in the cold season(hmm, needs work…). Sticking with this, at the equator, the mean daily high temperature varies from 110ºF to 105ºF. As you can see, my numbers are a little cockeyed at present, but I think with a bit of jiggering it could be made to work.
Of course, since I’m already insanely complicating your nice set-up, maybe proximity to the sea could moderate the temperatures and reduce the variations. Maybe within a maritime area the change in temperature due to latitude could be reduced by half and the range of temperatures over the year would be reduced by 5ºF. Maritime areas might be anything within 1000 miles of the coast at latitudes less than 30º(north or south), say 250 miles between 30º and 60º and no such thing between 60º and the poles.
Now, basing the seasonal variation base on the midrange mean daily high, let’s see what we get. For 45º, we get a midrange of 55ºF with a variation of +-25ºF.
If we reduce that variation by 1ºF for each degree of latitude down to 30 degrees then we get a variation of only +-10ºF at 30º latitude and the variation goes to zero at 20º latitude(I’ll have it stay at zero thereafter). That’s not a bad approximation of the tropics, so good enough. If we vary the midrange temperature the same way you do, we get an equatorial temperature of 85ºF, still missing, but not too badly. At the poles we get temperatures ranging around -20ºF, from 50ºF to -90ºF, which is excessive.
Eh, I thought it was worth a try…
I’ve got a fairly realistic (read that as acceptable for my purposes) system that is far to complicated to be useful, that takes into account: altitude, elevation, season, proximity to large water masses, etc. It’s been gathering dust for years due to its bulkiness. Sadly, I’ve never been able to come up with a simply system that doesn’t fall apart at the extremes.
The above works nicely for the mid latitudes but is a little extreme at the poles and the equator. Instead of adding complication to try to correct this, I actually went the other way. Using the above model, I adjusted the base temperatures for 15 and 75 degrees. The I use the 15 degree list for all latitudes between 0 and 30, the 45 degree list for latitudes between 30 and 60, and the 75 degree list for latitudes between 60 and 90. I realize that it is completely unrealistic but it lops off the extremes and more importantly, it is extremely simply (so much so that it may actually get used instead of sitting in a box with my last attempt.) :)
This is one of those things that fascinate me as an exercise, but I’m not sure how much it adds to the game. Needless to say, I’ll probably revisit the subject when I new approach occurs to me.
Climate modeling can get really complicated really fast. The more I play with it the more I realize that adding complexity tends to just make the model less realistic until you have something completely cumbrous, like an EBM or (ZOMG) a GCM! I have fun with this kind of stuff, but, honestly, I don’t think the gaming advantages are worth all that effort.
I believe weather can add a lot to the gaming experience if you play it right. First thing is, ditch the simulation. In the world-building stage I’m obviously all for figuring out the climate. That tells you what kind of environment you have and also what sort of weather makes sense. But day-to-day weather should fly at the speed of plot. A miraculous rainstorm in a desert can save the character’s lives(a little deus ex goes a long way…).
Weather is a good way to quietly and realistically railroad the story when necessary. Sure the characters can leave that odd village any time they like. Having the roads washed out in a torrential monsoon might convince your players to spend the night at that village and get involved in your carefully crafted scenario. At least after they’ve spent a few hours slogging through hip deep mud, getting attacked repeatedly by slimy things driven out of the ground by the rising water table and generally making no great progress.
Even without Saruman helping, a good hard blizzard hitting the passes might convince any adventurers trying to cross the Misty Mountains to look for another route…
Before steam(and even after, to a great degree), sea travel was all about the weather. Expeditions have been destroyed by starvation by getting caught in the doldrums at the wrong time(maybe not the most exciting adventure…), and The Perfect Storm could play hell with any voyage(read the book, I have no idea about the movie).
I hope at some point you post your, “acceptable for… (your)… purposes,” model. I find these things entertaining to a degree well beyond their actual usefulness.