about the author
warning: non-mainstream opinion

Are there unpredictable sudden thresholds?

Given that the actual data do not give any base for panicking, alarmists have to rely on predictions that either all this will accelerate very much in future, or that the horrible things will happen somewhere in the future in a quite unpredictable way. If some acceleration is claimed, ask for the evidence and, if it is presented, look at the resulting numbers. Say, the rather unimpressive actual rise of the sea level of 3 mm per year is claimed to accelerate, but the resulting predictions, around 3 cm per year, even if much larger, remain not very horrible.

The alarmist claims considered here are of the latter type. They predict that over a quite long period of time, nothing horrible happens. But then, quite sudden and unpredictable, the horror starts. The common sense reaction to such claims is, of course, to look how those researchers predict those unpredictable things.

Thresholds for ecosystems predicted by space-for-time substition

Here is how one alarmist has tried to use this technique to raise panic:

In Science Magazine, issue published 14/2/2020, we read that there are apparently three major thresholds in increasing aridity, (at .54, .7, and .8 iirc), after which crossings the entire supported ecosystem changes abruptly - in soil biome and chemistry, vegetation cover, faunal community, etc - but between which the changes are small and incremental.

The bad news is that about 20% of the land surface of the planet will likely cross one or more of those thresholds before 2100 CE - that's the AGW researchers's data supported most likely prediction, given current trends - and there will be little warning - the thresholds are apparently fairly sharp. [from here, emphasis mine]

The paper I have identified as Berdugo et al. 2020.

The method which is used to predict those sudden changes is named "space-for-time substitution:

It is worth noting that the results of this exercise are temporal extrapolations of results obtained using spatial gradients, and therefore constitute what is known as a space-for-time substitution approach (117). The use of this approach is common in ecology (118–120), as it allows to infer hypotheses related to temporal changes when temporal series are not available or do not suffice to cover processes that operate at very slow temporal rates (117, 119).
Berdugo et al. 2020, suppl. mat.

This is a quite reasonable method. How it works? We have some prediction about how the key variables like temperature and precipitation (or whatever matters most) will change in time. We have a particular place, and want to predict what will happen at that place in future because of climate change.

To find out what will happen we simply try to find another place, a place which actually now has the temperature and precipitation (or whatever matters most) which are predicted for the future of our starting point. So, we substitute the actual situation of another place in space for the situation expected to happen in the future of our initial point, we substitute space for time.

So this method is simple enough to be used by laymen too.

How laymen can use this method

The advantage of this method is that this is something which can be done by laymen too. Climate change predicted a temperature increase of 2 degrees. What will happen at your home town? Find out the average temperature of your home town. Then add these two degrees, and try to find a place which actually has this average temperature of two degrees more. Then you can look yourself how this looks like.

In the mountains this can be done on quite small distances, given that the average temperature depends on the altitude - we know very well that in the mountains the weather is colder. Moreover, this temperature dependence is the same everywhere, and the formula for this, which holds for altitudes up to 11 km is quite simple: The temperature does down by 1 degree C per 150 m of higher altitude.

So let's see how this would apply to the particular place in the picture. We see here three different ecosystems: The highest one, which survives the lowest temperatures, is simply grass. Below this, there are bushes. And down in the valley we can see wood.

Of course, there is also another factor beyond temperature which is important in the mountains - exposure to wind. If a place is better protected from wind, plants can survive even if the temperature is lower. But if one compares the places where one would not expect that the average wind will be different, as on the other side of the lake, the border between grass and bushes is almost exactly a straight line defined by the same altitude.

The border between the wood and the bushes is similarly a quite straight line, and it has even a well-established name: timber line or tree line.

To apply the space-for-time substitution, let's see how one can predict what happens at this place if the temperature increase is two degrees of Celsius in 100 years. Given the relation of 1 degree C per 150 m, this translates into 300 m of altitude, in the direction down the hill because in this direction it becomes warmer. So, go down until you reach an altitude 300 m below this point and you will see how the place on the picture will look like in a century.

We can also predict, with the same method, when what happens. Say, the prediction is a 2 degrees increase in 100 years. Then, if the increase is linear, that means 1 degree in 50 years or 0.2 degrees in ten years. This translates into 150 m corresponding to 50 years and 30 m corresponding to 10 years, and 3 m to 1 year. You want to know when the bushes will be replaced by wood? Find out the altitude of the actual position, then go down to the timber line - the altitude where the bushes end and the wood begins. Find out the altitude of that place. Compute the difference in meter, divide the result by 3 and you have the number of years you have to wait until it is warm enough for wood to grow at your starting point.

Unpredictable thresholds?

So, some aspects of the prediction of thresholds are correct: At a given place, there will be long periods of time when nothing changes - bushes remain bushes. Then, quite sudden, the timberline reaches that place, and the bushes will be replaced by wood. After this, again, nothing changes at that place.

But, different from the alarmist claim, the "sudden change" can be predicted, and can be predicted quite easily by looking around where the timber line is actually located.

So, in this case it appeared that the "unpredictable Threshold" was quite predictable, and the method of prediction is accessible to laymen too, if they simply follow common sense.


  1. Berdugo M. et al. (2020). Global ecosystem thresholds driven by aridity. Science 367, 7877-790; see also supplementary materials.