20 views
Skip to first unread message

Mats Almgren

unread,
Nov 8, 2020, 10:30:44 AM11/8/20
to Anastassia Makarieva, Biotic Regulation of the Environment
Dear Anastassia,

The old thread is rather unmanageable, so I start a new one. Here I will comment on some of your comments.
1. Ruttenberg's review shows that dust flux is larger than oceanic flux of P. The same magnitude is discussed by . But that does not mean that the oceanic flux (which is about an order of magnitude lower) is not enough for the ecosystems.
 I am not quite certain of the meaning of this comment. I do understand, however, that the amount of a nutrient like P that is circulating within an ecosystem is independent of the flows of P in and out, as long as the two are equal. But the rate of change of the amount of P in the system (and I guess the rate of change of the living biomass) will depend on magnitudes of these flows. And if the inflow is small, the outflow must be equally small.
2. Studies of the Hawaiian ecosystems that I referred to do not assess current carbon acquisition by these ecosystems (cf. Fig. 4 lowest panel of Vitousek et al. 1997). They just compare current stores of soil carbon among sites with different ages and infer long-term carbon accumulation rates that have nothing to do with anthropogenic CO2.
Yes, you are right, I misunderstood fig. 4F. But I have now found some studies about the Amazonian rainforest: Carbon Dioxide Uptake by an Undisturbed Tropical Rain Forest in Southwest Amazonia, 1992 to 1993, John Grace, et.al Science 270 pp. 778-780, 1995. There was a net uptake at that time. But now it seems as if the system is close to a tipping point, according to Luciano Gatti and Carlos Nobre (https://thebulletin.org/2020/05/is-amazon-rainforest-going-from-carbon-sink-to-carbon-source/). Even those parts of the forest that are not destroyed by logging or fire suck up much less carbon dioxide now. This is rather scary. If the biotic regulation stops working (due to the warming?) – what could save us?
3.An increase in primary productivity after adding something to the ecosystem is not an indication that this "limits" the ecosystem functioning. It means that the ecosystem reacts to this disturbance by higher NPP (which is not the only possible reaction). It is like fever in an infected human body. Metabolic rate does rise, but that does not mean that lack of infection limits human energy turnover.?
4. Long-term fertilization experiments are few, but they show that if the disturbance persists, the ecosystem may begin to disintegrate e.g. by losing soil carbon (see e.g. Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization / M. C. Mack,

I know that this is the view according to BRT, and I am not surprised that long term fertilization may disrupt the soil. But what is clear from the Hawaiian experiments is that the response to addition of N or P is different in old soil (rich in N, poor in P) than in young soil (poor in N, rich in P).  According to BRT the response to a perturbation should always be according to le Chatelier's principle, i.e. such that the perturbation is reduced. In both old and young soils, growth is the response to addition of the ”limiting” nutrient, but if the ”limiting” one is not altered, the response is something else that presumably reduces the availability of the added nutrient. The local ecosystem must be different, with different preferred levels of N and P. Since the macro vegetation did not change from one system to the other, it suggests that the microorganisms are different. I have looked for ”below-ground biodiversity” studies, but have not found any for the Hawaiian systems. But there are a lot of other studies, and some of the more recent ones show that the below-ground biodiversity may change substantially even if the above ground biodiversity remains the same.(https://www.pnas.org/content/116/14/6891) Thus, the ecosystem evolves, adapts, in response to the changing conditions. This is what one would expect, from a normal Darwinian perspective, but how about BRT?

Best regards
Mats

Anastassia Makarieva

unread,
Nov 9, 2020, 12:12:11 PM11/9/20
to Biotic Regulation of the Environment
Please see my replies in green below your comments

1. Ruttenberg's review shows that dust flux is larger than oceanic flux of P. The same magnitude is discussed by . But that does not mean that the oceanic flux (which is about an order of magnitude lower) is not enough for the ecosystems.
 I am not quite certain of the meaning of this comment. I do understand, however, that the amount of a nutrient like P that is circulating within an ecosystem is independent of the flows of P in and out, as long as the two are equal. But the rate of change of the amount of P in the system (and I guess the rate of change of the living biomass) will depend on magnitudes of these flows. And if the inflow is small, the outflow must be equally small.

"And if the inflow is small, the outflow must be equally small." -- true. The problem is that most land is now disturbed and so it makes it difficult to estimate the characteristic rate of phosphorus loss from an intact ecosystem. From Table 2 of the review it is clear that river runoff of particulate P approximately coincides with mined P: all fertilizers that we mine end up in the ocean. But in intact ecosystems there is no loss of particulate organic matter (clean waters). E.g. in the Amazon the runoff appears to be dominated by soluble P. From Table 2 it is clear that the outflow of soluble P is an order of magnitude lower than particulate P and that this outflow coincides in the order of magnitude with the inflow of dust P.

However, soluble P outflow from disturbed lands can also be much higher than it would be from intact land. If the intact outflow an order of magnitude lower than it is now, that would coincide with the inflow of ocean-derived P. This inflow is one order of magnitude smaller than from dust.

An indication that this logic is reasonable is provided by the study of Swap et al. 1992 (page 146) where they say that the hydrological loss values of phosphorus from the Central Amazon Basin "vary from an order of magnitude less than our deposition values to the same order of magnitude". (where by deposition they mean dust derived P)

Thus I think land biota could have evolved relying predominantly on the oceanic P.

 
2. Studies of the Hawaiian ecosystems that I referred to do not assess current carbon acquisition by these ecosystems (cf. Fig. 4 lowest panel of Vitousek et al. 1997). They just compare current stores of soil carbon among sites with different ages and infer long-term carbon accumulation rates that have nothing to do with anthropogenic CO2.
Yes, you are right, I misunderstood fig. 4F. But I have now found some studies about the Amazonian rainforest: Carbon Dioxide Uptake by an Undisturbed Tropical Rain Forest in Southwest Amazonia, 1992 to 1993, John Grace, et.al Science 270 pp. 778-780, 1995. There was a net uptake at that time. But now it seems as if the system is close to a tipping point, according to Luciano Gatti and Carlos Nobre (https://thebulletin.org/2020/05/is-amazon-rainforest-going-from-carbon-sink-to-carbon-source/). Even those parts of the forest that are not destroyed by logging or fire suck up much less carbon dioxide now. This is rather scary. If the biotic regulation stops working (due to the warming?) – what could save us?

Yes the Amazon forest has been dramatically disturbed by cutting and human-induced fires. Please see these presentations by Antonio Nobre https://youtu.be/3vHfVf-9YqY and Juan Salazar https://youtu.be/EEvEBLYycR0 Yes it is scary. If land biota is destroyed, our only hope will be that the oceanic biota could manage the Earth's system for some time alone.
 
3.An increase in primary productivity after adding something to the ecosystem is not an indication that this "limits" the ecosystem functioning. It means that the ecosystem reacts to this disturbance by higher NPP (which is not the only possible reaction). It is like fever in an infected human body. Metabolic rate does rise, but that does not mean that lack of infection limits human energy turnover.?
4. Long-term fertilization experiments are few, but they show that if the disturbance persists, the ecosystem may begin to disintegrate e.g. by losing soil carbon (see e.g. Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization / M. C. Mack,

I know that this is the view according to BRT, and I am not surprised that long term fertilization may disrupt the soil. But what is clear from the Hawaiian experiments is that the response to addition of N or P is different in old soil (rich in N, poor in P) than in young soil (poor in N, rich in P).  According to BRT the response to a perturbation should always be according to le Chatelier's principle, i.e. such that the perturbation is reduced. In both old and young soils, growth is the response to addition of the ”limiting” nutrient, but if the ”limiting” one is not altered, the response is something else that presumably reduces the availability of the added nutrient.

From the BRE point of view, it is possible that on young soils NPP is not enhanced by the addition of phosphorus, because phosphorus is abundant and already being removed from the ecosystem at a maximum possible rate.
 
The local ecosystem must be different, with different preferred levels of N and P. Since the macro vegetation did not change from one system to the other, it suggests that the microorganisms are different. I have looked for ”below-ground biodiversity” studies, but have not found any for the Hawaiian systems. But there are a lot of other studies, and some of the more recent ones show that the below-ground biodiversity may change substantially even if the above ground biodiversity remains the same.(https://www.pnas.org/content/116/14/6891) Thus, the ecosystem evolves, adapts, in response to the changing conditions. This is what one would expect, from a normal Darwinian perspective, but how about BRT?

As before with Figure 4F of Vitousek et al. 1997, chronosequences do not tell much about evolution -- they all describe extant ecosystems. In the study you referred to data are presented concerning the dominant species of soil bacteria, fungi, invertebrates. This is obvious that as the soil substrate ages, the local ecological community has to employ different parts of its machinery to do environmental regulation. No new species have evolved, they are all nearly ubiquitously present (at least the smallest ones see e.g. a review of "the ubiquity hypothesis" here https://onlinelibrary.wiley.com/doi/full/10.1111/mec.12507 ) but they are assigned different roles depending on the external conditions (in this case -- the age of soil substrate).

Best wishes,
Anastassia
 

Best regards
Mats


Mats Almgren

unread,
Nov 10, 2020, 10:50:07 AM11/10/20
to Anastassia Makarieva, Biotic Regulation of the Environment
Dear Anastassia,

It is not easy to change thinking-habits. Thank you for your patience!

1. Thus I think land biota could have evolved relying predominantly on the oceanic P.

Maybe, but dust was present when vegetation started on land, and has probably been present ever since.

2.Yes the Amazon forest has been dramatically disturbed by cutting and human-induced fires. Please see these presentations by Antonio Nobre https://youtu.be/3vHfVf-9YqY and Juan Salazar https://youtu.be/EEvEBLYycR0 Yes it is scary. If land biota is destroyed, our only hope will be that the oceanic biota could manage the Earth's system for some time alone.

Interesting presentations, thanks! And yes, the forests not directly perturbed were of course affected by the perturbation of the biotic pump.

3. As before with Figure 4F of Vitousek et al. 1997, chronosequences do not tell much about evolution -- they all describe extant ecosystems. In the study you referred to data are presented concerning the dominant species of soil bacteria, fungi, invertebrates. This is obvious that as the soil substrate ages, the local ecological community has to employ different parts of its machinery to do environmental regulation. No new species have evolved, they are all nearly ubiquitously present (at least the smallest ones see e.g. a review of "the ubiquity hypothesis" here https://onlinelibrary.wiley.com/doi/full/10.1111/mec.12507 ) but they are assigned different roles depending on the external conditions (in this case -- the age of soil substrate).

Yes, I do understand that these ecosystems are contemporary. But if there were systematic differences between them, it should imply that they had evolved differently. If the ubiquity hypothesis is correct, this simply means that different species dominate, depending on the environmental conditions – which is also what BRE says!  As I understand it from the linked article, the ubiquity hypothesis will probably not be proven any time soon, so it could be either way. – For prokaryotes it is rather hard to distinguish different species, since they seem to exchange genetic information rapidly and freely, and also pick up pieces from viruses. The genetic code should be there, but maybe not in the same species in all systems.  

Best regards
Mats


--
You received this message because you are subscribed to the Google Groups "Biotic Regulation of the Environment" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biotic-regulat...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/biotic-regulation/CAKz3-9-ZKAYvCqL6WWa%2Bz7-NdJ7zEgkUoJr%2B5BxM-SjPQ27euw%40mail.gmail.com.
Reply all
Reply to author
Forward
0 new messages