Can living things "adapt" when the temperature changes?
Run the experiment on this page ↓
Short answer (general ecology)
An individual does not change its inherited traits to suit a new environment. A population can adapt over generations when individuals that happen to suit the new conditions leave more offspring. That takes heritable variation and time; when the change is too large or too fast, a population can decline or die out instead.
What you can do here
Compare two herbivore populations whose individuals differ in the temperature that suits them, a trait passed from parent to offspring. In one world the temperature rises at a set time. Follow the count and the mean temperature optimum of the population.
Limit
This is an abstract educational model. It does not forecast what would happen in any real ecosystem.
Run the experiment
- Predict
- Watch one run
- Compare 5 runs
- Read the result
After the temperature changes by +150 in B, how do you think the mean temperature optimum of its population will move compared with A?
You can start without choosing. Your choice is not sent anywhere.
Watch one run
tick 0 / 0Preparing the world…
A: nothing changed
B: temperature changed by +150
| A | B | |
|---|---|---|
| Herbivores | ||
| Grass | ||
| Temperature |
Keyboard, while the focus is in this panel: Space plays and pauses, → advances one step, R restarts, 1 and 2 show A and B.
Selected individual
Click or tap an individual in either world to see its values here. The button under each world selects one without a pointer.
What the shapes and colours mean
Body colour = temperature optimum
Bluer bodies do best in the cold and redder ones in the heat; those in between are off-white. The number is the temperature at which the individual does best. The band around the edge of the world shows the world’s temperature right now, on the same colour scale: the closer a body’s colour is to the band, the better that individual suits the current temperature.
Ground colour = amount of grass
The amount of grass is shown in three steps: the colour of soil where there is almost none, light green where there is grass, and deeper green where it is thick.
Dullness and thinness = reserves
As reserves run low, the body turns duller and thinner. An individual whose reserves run out dies.
Size = growth
Newborn individuals are drawn small and grow to adult size.
Compare repeated runs
One run can differ from the next by chance. The same comparison is repeated with different random seeds to see how consistent the difference is.
Try it: change a condition and compare again
Write down what you expect first. The random seeds stay the same, so a difference from the standard result comes from the setting you changed.
How far the temperature moves in B. A negative number lowers it; 0 means nothing changes. Standard: +150. Range: -300 to +300.
0 changes the temperature at once. A larger number spreads the same change over that many ticks. Standard: 0. Range: 0 to 2,000.
How much individuals differ at the start. 0 means all start with the same optimum. Standard: 35. Range: 0 to 100.
The chance, at each birth, that the offspring’s optimum differs slightly from its parent’s. Standard: 5%. Range: 0% to 20%.
Saved results
Results are stored only inside this browser and are never sent to a server. What is stored is the settings, the random seeds and the summary numbers for each replicate; the tick-by-tick record is not stored, because the same settings always reproduce the same result. Up to 50 results are kept across all experiments and the free lab, and the oldest are removed beyond that. They do not carry over to another browser or device: use a share link to move one. Deleting cannot be undone.
Move saved items to another browser
Everything saved here (all experiments and the free lab) can be written to one file. Load that file in another browser or device and the same items are listed there. Loaded results use the numbers in the file as they are; run the settings again if you want to check them. The file stays on your device and is not sent anywhere.
Reference result (20 replicates)
These values were computed in advance with the same engine that runs in your browser. Running the same settings and seeds yourself gives exactly the same numbers.
A and B compared
| Measure | Result |
|---|---|
| Lowest herbivore count (after the intervention) | B was lower in all 20 replicates (median A 307, B 60). |
| Herbivore count (mean over the observation window) | B was lower in all 20 replicates (median A 368, B 354). Note that the difference is small: the median difference is under 10% of A. |
| Shift in the mean temperature optimum (from the intervention to the end) | B was higher in all 20 replicates (median A -0.3, B 109.4). |
| Marker variants remaining (at the end) | B was lower in all 20 replicates (median A 6, B 1). |
Observed What was measured in these runs.Reading How the measurements can be read, given the rules of this model.
- ObservedIn 20 of 20 replicates, B (temperature changed) fell to a lower count than A. The lowest count in B was, at the median, 15% of the count at the intervention.
- ObservedIn the 20 replicate(s) where B did not go extinct, the count in B over the observation window was, at the median, 97% of A.
- ObservedIn 20 of the 20 replicates where neither A nor B went extinct, the mean temperature optimum in B moved in the same direction as the temperature change, and by more than in A (median shift: A -0.3, B 109.4).
- ReadingIn this model an individual's optimum never changes after birth. The mean can be read as having moved because individuals whose optimum was closer to the new temperature left more offspring (selection).
- ObservedEven in A, where the temperature was not changed, the mean optimum moved a little (median size 3.9; of 20 replicates, up in 9 and down in 11).
- ReadingThe movement in A had no consistent direction, so it can be read as change due to chance.
- ObservedIn 20 of the 20 replicates where neither A nor B went extinct, B also ended with fewer variants of the marker that does not affect survival or reproduction (median at the end: A 6, B 1).
Engine version 0.1.0, 20 replicate(s) with seed(s) 101–120. Observation window: ticks 3,000–3,600. These are observations from this model with these settings, not predictions about real ecosystems.
Charts
Show as a table
| Tick | A median | A range | B median | B range |
|---|---|---|---|---|
| 0 | 168 | 168–168 | 168 | 168–168 |
| 450 | 375 | 329–465 | 375 | 329–465 |
| 594 | 362 | 312–413 | 362 | 312–413 |
| 648 | 369 | 333–403 | 131 | 103–172 |
| 702 | 375 | 346–413 | 71 | 54–103 |
| 792 | 374 | 308–412 | 62 | 34–95 |
| 900 | 355 | 325–397 | 69 | 32–127 |
| 1,350 | 370 | 344–402 | 283 | 47–458 |
| 1,800 | 371 | 340–429 | 299 | 174–371 |
| 2,250 | 366 | 325–403 | 328 | 228–401 |
| 2,700 | 363 | 320–428 | 335 | 298–421 |
| 3,150 | 354 | 324–404 | 353 | 301–388 |
| 3,600 | 364 | 304–425 | 360 | 315–394 |
Show as a table
| Tick | A median | A range | B median | B range |
|---|---|---|---|---|
| 0 | 501 | 491–507 | 501 | 491–507 |
| 450 | 499 | 493–506 | 499 | 493–506 |
| 594 | 501 | 494–506 | 501 | 494–506 |
| 648 | 500 | 493–507 | 518 | 511–524 |
| 702 | 500 | 492–508 | 530 | 522–536 |
| 792 | 501 | 492–510 | 536 | 530–545 |
| 900 | 501 | 491–510 | 542 | 533–550 |
| 1,350 | 501 | 491–509 | 557 | 539–570 |
| 1,800 | 500 | 491–511 | 568 | 541–586 |
| 2,250 | 500 | 489–508 | 584 | 545–611 |
| 2,700 | 500 | 484–508 | 593 | 563–614 |
| 3,150 | 499 | 483–511 | 603 | 578–616 |
| 3,600 | 498 | 480–512 | 611 | 587–620 |
Show as a table
| Tick | A median | A range | B median | B range |
|---|---|---|---|---|
| 0 | 60% | 60%–60% | 60% | 60%–60% |
| 450 | 24% | 19%–27% | 24% | 19%–27% |
| 594 | 25% | 21%–30% | 25% | 21%–30% |
| 648 | 25% | 19%–32% | 34% | 27%–43% |
| 702 | 24% | 19%–31% | 62% | 52%–72% |
| 792 | 24% | 20%–30% | 87% | 81%–91% |
| 900 | 25% | 20%–28% | 87% | 78%–93% |
| 1,350 | 24% | 20%–32% | 37% | 20%–92% |
| 1,800 | 25% | 20%–29% | 36% | 29%–74% |
| 2,250 | 24% | 20%–30% | 35% | 28%–60% |
| 2,700 | 24% | 20%–29% | 32% | 22%–45% |
| 3,150 | 24% | 20%–30% | 29% | 23%–37% |
| 3,600 | 26% | 19%–32% | 27% | 21%–38% |
Show as a table
| Tick | A median | A range | B median | B range |
|---|---|---|---|---|
| 0 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 450 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 594 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 648 | 0.0 | 0.0–0.0 | 2.3 | 1.7–3.2 |
| 702 | 0.0 | 0.0–0.0 | 0.3 | 0.1–0.5 |
| 792 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.1 |
| 900 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 1,350 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 1,800 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 2,250 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 2,700 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 3,150 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
| 3,600 | 0.0 | 0.0–0.0 | 0.0 | 0.0–0.0 |
How the spread of temperature optima moves
Each row is one moment. The shape shows how many herbivores have each temperature optimum, and the vertical mark shows the temperature of the world at that moment. Watch whether the shape in B moves towards its mark, and how long it takes. The 20 replicates are added together. Each replicate settles at a slightly different place, so the combined shape can show several peaks.
Show as a table
| Tick | A: temperature | A: middle of the distribution (approx.) | B: temperature | B: middle of the distribution (approx.) |
|---|---|---|---|---|
| 600 | 500 | 504 | 500 | 504 |
| 1,100 | 500 | 497 | 650 | 551 |
| 1,600 | 500 | 504 | 650 | 559 |
| 2,100 | 500 | 497 | 650 | 583 |
| 2,600 | 500 | 497 | 650 | 598 |
| 3,100 | 500 | 497 | 650 | 598 |
| 3,600 | 500 | 497 | 650 | 614 |
Printed from https://ecosimarium.com/experiments/climate-shift/