First of all, I am tagging this with the because this phenomena is not a result of something natural.

What would happen to a whole forest region if a patch of desert magically appeared at its center? Below is an example of what I am trying to illustrate.


  • The beige-coloured parts represents the desert that was magically displaced in the forest-y region
  • Whole forest = 25 mile radius
  • The climate is not transported
  • The desert is a Sand desert

enter image description here

Here is an image to illustrate the elevation.

enter image description here

As you can see in the image, it is like the center of the forest was cookie-cutted and then replaced with this patch of desert. And also, the elevation would be flat.

Question: What would happen to the forest (by forest, I mean the forest with the desert in it) within 1 year and within 5 years?

I am mostly curious whether the forest will turn into a desert or the desert will turn into a forest.

With my proposition, let us assume that the only applies to the magical translocation of the said desert and nothing else. The forest is just a normal forest.

Thank you.

EDIT: This is a little off-topic, but this ought to help you guys for generating maps. If anyone is interested, this is the tool I used to generate this map. This tweet also contains excellent resources for generating something like this.

  • 4
    $\begingroup$ What is the elevation of the desert part relative to the terrain it replaced ? Note the rivers on your map - the water comes from altitude so if you remove the high ground that's a significant change for water flow. How big is the area ? There's a big difference between a ten mile radius and a thousand mile radius for climate purposes. $\endgroup$ Apr 3, 2018 at 2:27
  • 25
    $\begingroup$ A desert is a place where the amount of rain is less than 300 mm per year. You cannot "move a piece of a desert" -- what you move is a piece of sandy or rocky terrain, which will no longer be a desert if you move to a position where it rains more. So what you are really asking is what happens in a large forest if you cut all the trees in a 25 mile patch: the answer is, grass will grow on it, then trees. And 5 years is not "long term"; when speaking about forests, "long term" at least the time needed for two or three generations of trees, so that the forest can reach equilibrium. $\endgroup$
    – AlexP
    Apr 3, 2018 at 4:09
  • 1
    $\begingroup$ "The elevation is flat." The shadows in your picture indicate hills. The fact that rivers radiate outwards also means that there are central hills. Specifically, a central spine running from SW to NE. Thus, the forest can't be flat. $\endgroup$
    – RonJohn
    Apr 3, 2018 at 4:48
  • 2
    $\begingroup$ Also, plateaus are flat, and deserts can be at any altitude. All in all, this is a very poorly thought question. $\endgroup$
    – RonJohn
    Apr 3, 2018 at 4:49
  • 1
    $\begingroup$ Is the climate also magically moved to the forest? Or just the sand? $\endgroup$
    – Rolf ツ
    Apr 3, 2018 at 7:29

6 Answers 6


What happens largely depends on the climate-conditions it is set in.

Most important question: How does this affect Rainfall.

You cut back the Mountain, which could result in less rain because of the missing elevation. If the cross section is to scale, you cut back about 2.5 miles of mountain, which is a lot. It will considerably lessen the rainfall and allow for faster winds on the leewards side. See how this works here

You also eliminated the trees, so the days will become hotter and water will evaporate faster. Trees reflect infra-red and protect the humid air below them from getting blown away. Some historic examples of islands suggest that deforestation can lead to a change in local climate to a more barren one. See Easter Islands for example

Sand heats up much faster in the daylight and cools of more quickly in the night. The turbulence by the hot air from the desert could result in some local thunderstorms. Similar to urban heat island effect

Then it depends how windy the whole region is, and if there is a main direction the wind is coming from.

  1. Wandering dunes: Sand will probably get blown into the forest and could bury it all the way to the coast, in the main direction of the wind. See Rubjerg Knude

  2. Depending on rainfall, the vegetation will creep into the desert, or the desert will expand.

  3. Lakes and streams will appear in the desert, providing some early oasis, if there is still sufficient rain.

Also 1 Year is not a long time, you´d probably only have some strange weather and some new rivers. I tried to imagine 5 years with a dominant east wind:

enter image description here

  • 1
    $\begingroup$ What "hot air from the desert" I don't see where you think this is coming from? he's trans-located a patch of sand & any hot hair that came with it during the initial trans-location will dissipate within a matter of hours if not sooner, sand doesn't magically manufacture heat & he's clearly stated in the question that the only magic involved is the initial trans-location of material. $\endgroup$
    – Pelinore
    Apr 3, 2018 at 17:33
  • 15
    $\begingroup$ @Pelinore Assuming enough sunshine, sand is going to heat differently than forest. With sand the assumption is it's going to be similar to an urban heat effect. $\endgroup$
    – Michael
    Apr 3, 2018 at 17:44
  • 3
    $\begingroup$ @Michael nuff said, I can accept that, the answer could perhaps profit from a nod at this cause in the text to forestall responses like mine :) $\endgroup$
    – Pelinore
    Apr 3, 2018 at 17:55
  • 3
    $\begingroup$ "You also eliminated the trees, so the days will become hotter and water will evaporate faster." This is incorrect. Trees cause water to evaporate faster. That may seem intuitive, but consider the difference in surface area of exposed water between flat sand and every leaf on every tree in a forest. Rain-forests raise the local humidity substantially. They literally create their own rain. $\endgroup$
    – Necoras
    Apr 3, 2018 at 19:39
  • 1
    $\begingroup$ @Pelinore: Look at day/night Tempratures in the sahara. Trees reflect infra-read and isolate the ground from cooling off at night. They provide a dampening effect on temperature swings. $\endgroup$
    – Daniel
    Apr 3, 2018 at 21:07

The desert will slowly be overgrown by the forest.

Desert are created from a lack of rain and looking at the map, rain isn't an issue.

Plants will start growing in from the edges and more hardy plants will pop up in the middle.

Leaf litter will build up. Animals will move in and eventually everything will be as before.

As for time frames, there is no scale on the map. Is it 12 square miles on an island or 12 million square miles on a continent?

  • $\begingroup$ I updated my question. Let us assume this is 25 mile radius. $\endgroup$
    – Bwrites
    Apr 3, 2018 at 3:07
  • 3
    $\begingroup$ The desert part could even be so low, that the water from the mountains could create a lake. $\endgroup$ Apr 3, 2018 at 5:04
  • 4
    $\begingroup$ Elsewhere, OP states that the desert is sand all the way down, which I think changes this answer. This isn't just a recently-desert chunk of land. So while leaf litter will allow for plants in the long term, in the short term, there's no means of water retention in the desert portion to allow for takeover. $\endgroup$
    – Carduus
    Apr 3, 2018 at 14:25
  • 1
    $\begingroup$ I don't see how the forrest will survive long enough to cover the desert in soil so that plants can grow there. The amount of water keeping the forrest alive is drastically cut when the desert appears. $\endgroup$
    – gibson
    Apr 4, 2018 at 11:58
  • 1
    $\begingroup$ @CharlesBurge A key flaw in your comparison there is that forest fires cover the ground in incredibly fertile ashes which allow rapid growth and recovery of plant life. $\endgroup$
    – Orgmo
    Apr 4, 2018 at 12:01

I'm assuming this is an island in an ocean and not in an inland lake. Given that, there will probably be a fairly consistent wind pattern and it will have plenty of water in it. (Otherwise there would have been no pre-magic forest.)

The picture shows a mountainous interior, but also seems to show the forest to be pretty uniform. That says that the original mountains were not very high. If the mountains had been higher, they would have cast a major rain shadow and the leeward side of the island would have been dry. (On the Big Island of Hawaii, the windward side gets as much as 20 feet of rain a year, while the leeward side has sections which get ten inches and look like the dryer parts of Wyoming.)

(N.b., if the island was a lot larger and the desert correspondingly bigger, the lack of evaporation from the missing forest would have an effect, and would somewhat decrease rain downwind. At 25 miles this can probably be ignored.)

Your picture shows sand "all the way down". This would have a profound impact on the post-magic development. Because the mountains were low, their removal would have little effect on the weather patterns. So we can expect the rain pattern would be about the same as it was pre-magic.

The deep sand would greatly reduce the speed at which regrowth of forest takes place because there would be no water at all in the top layers -- rainfall would very quickly percolate downwards to the level of the water table which would probably not be that much higher than sea level -- it all depends on how impervious the deeper layers of earth under the original forest were. If the water drains away, it is nearly impossible for soil to develop because organic matter that gets blown in by the wind says dry and gets blown away again. (You see this in the Pumice Desert next Crater Lake in Oregon or in Desert of Maine.)

So recovery would be slow. In fact, if the border of forest was narrow enough, the deep sand would lower the water table under the forest and might actually kill off the forest. My gut feeling is that this would not happen with a ratio of forest to sand as shown.

  • 2
    $\begingroup$ Welcome to the site, Mark! +1 for the good answer, but I'll also point out that the community tends to shy away from "gut feeling" answers. if you could do a little research (and include it here) to confirm or refute your suspicions, this would be an even better answer. $\endgroup$
    – Frostfyre
    Apr 3, 2018 at 12:39
  • 2
    $\begingroup$ Sadly, this isn't the sort of thing that people do much research on! I'll happily delete the "gut feeling" sentence if that would help. (Equally, I could add a plausible argument, but it would be handwaving that's basically the same thing in disguise. Let me know what community standards prefer. $\endgroup$
    – Mark Olson
    Apr 3, 2018 at 13:22
  • 2
    $\begingroup$ It should be fine as is. Still, I wouldn't be surprised to discover the research exists somewhere on the Internet. It's a pretty big place, after all. :) $\endgroup$
    – Frostfyre
    Apr 3, 2018 at 14:19
  • $\begingroup$ I must have missed the part of Wyoming that is covered in a'a. $\endgroup$ Apr 4, 2018 at 20:57

This is actually a much more interesting question than it looks like on its face. So now we have two 'camps' warring against each other.

Because of the apparent proportion of forest to desert, the immediate effect is going to be a drop off in the humidity levels of the remaining forest, as it lost its most humid portion. This is going to slow vegetation growth and mess with animal symbiosis something fierce. Likewise, the fact that the desert sands send hard rains flooding across its surface, and softer rains just sink deep in below where a tree could get it, it would take generations to build up enough tilth to sustain forest plants. So this sudden desert has a very good chance of messing up the existing forest.

The answer? Eventually, this whole island is going to be savannah/grasslands. The forest animals are very likely to compete over suddenly scarce resources and die out, and those that are left are very likely to kill off most of the larger plants in a desperate attempt at survival. Grasses don't need complex ecosystems to maintain, and are likely to exist where the forest meets the ocean anyways. So slowly, the desert is colonized by grasses, and then the scraggly bits of what's left of the forest will be colonized too.

  • 3
    $\begingroup$ This is my favorite answer. BUT... savannahs don't get a ton of rain and this place doesn't look like it has a rain problem. That might change things? $\endgroup$
    – Len
    Apr 3, 2018 at 19:48
  • 1
    $\begingroup$ I don't mean savannah in the steady biome sense, but in the grassland-with-trees sense. It would be a forest -> light forest -> savannah -> temperate grasslands. In order to maintain a savannah, you're correct; you'd need 20-50 inches of rain concentrated in a single season. $\endgroup$
    – Carduus
    Apr 3, 2018 at 19:58
  • $\begingroup$ Where is the grass seed going to come from? Transported over with the sand (possible) or blown in from the margins of the island (probable)? Because they certainly don't exist in the center of a tropical rain forest. $\endgroup$
    – Jason K
    Apr 4, 2018 at 13:13
  • $\begingroup$ As I mention in my answer, >Grasses don't need complex ecosystems to maintain, and are likely to exist where the forest meets the ocean anyways. $\endgroup$
    – Carduus
    Apr 4, 2018 at 15:45

Let's start with the actual example of the artificially generated desert that exists in the real world, called Błędowska Desert. There are some differences compared to your question but it can give you an idea what will happen since there is no change of the climate. As you can see now the dessert is slowly being grown over and that's what most likely will happen in your world's case. Since you do not change the elevation, the speed of this process most likely will be much faster since the conditions will be much more in favour of the flora restoration. So in 1 year the changes will be probably visible and in 5 year serious (for instance there'll be no longer purely-sand places), however most probably the forest will not be fully back in it's place. Also the increased amount of sand might change the type of initially dominating trees to those having more extensive roots system.


I've missed the original forest line/elevation which means the processes of restoring the flora will be slow, especially initially. Also there are some springs since there are rivers starting in the affected area. We may expect those springs remain so the water supply remain provided, but the river beds will probably change (depending how deep the sand layer is). So the desert will be grown over from edges and plants will show in the areas most humid. Anyway changes within 5 years are going to be rather small and noticeable mostly in the areas where water can be found. In longer run (decades) the dessert will decay and eventually get completely overgrown. The rest of my answer remain fully valid.

  • $\begingroup$ Welcome to WorldBuilding Ister! If you have a moment please take the tour and visit the help center to learn more about the site. Have fun! $\endgroup$
    – Secespitus
    Apr 4, 2018 at 11:32
  • $\begingroup$ But the elevation is changed. The entire desert area is flat. $\endgroup$
    – gibson
    Apr 4, 2018 at 12:02
  • $\begingroup$ Thanks @gibson for pointing it out. I've missed the old forest line. Answer updated accordingly. $\endgroup$
    – Ister
    Apr 5, 2018 at 8:17

As others said, it depends on weather conditions - maybe the forest will retake the land, maybe the desert will grow.

One theory about Sahara desert for example is that the location was relatively fertile and live (savannah-like?) during Ancient Egyptian times, but considerable infestation by Roman legions who wiped out a patch of clean land (no trees - good visibility and protection from night raids) to secure a camp for several thousand people here and there and everywhere over several centuries, combined with not so much rainfall (slow regrowth) and frequent strong winds to carry away the rootless land and cover nearby (not yet cut down) live lands with dead sand, ended up with the huge desert we know today.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .