The higher the understanding the higher the accuracy there are not absolutes

The function of benthic ecosystems might be easier to accurately measure than the complex interactions between taxa! It's far easier to observe outcomes than to explain why those outcomes occurred. That said, we know far more about organisms in isolation than as a consortium, but I've seen a recent emphasis on understanding those associations better. This should help us understand how disturbances will affect benthic ecosystem function.

If accuracy is ability to predict and model processes in a wholistic way, it will certainly depend how much we know of individual compnents, and I doubt we can do that with the level of understanding we have (case in point how much/little we know of nematode-related processes). On the other hand, we can make accurate predictions of specific processes based on the limited information we have.

A question for the Nobel Prize, eh?! ;) What do you mean under 'benthic functioning'? First, at 'better-or-worst' level, one could use different biotic (Woodiwiss etc) or diversity (Shannon etc) indices. Sometimes structural parameters (eg predators ratio) could say something.

Second, we could study productivity of most abundant species and thus determine energy flows within and between communities.

Third, as we know main features of communities in different conditions, we could generally predict changes in species composition and diversity the taken community will undergo following this or that changing (eutrophication, salinization, pollution, etc).

In my opinion, the diversity of communities isn't unlimited, and if we made the typology of them we could predict their 'evolution' accurately. Unfortunately, we're far from this.

We need not only information about the interactions between taxa, but also about the interactions between each taxa and different environmental factors. Only the study of both things at different temporal and spatial scales will lead us to an accurate understanding of benthic ecosystems functioning. So, when doing predictions or extracting conclusions we must be very cautious. Anyway, we already know many things on this subject, a huge insight on this topic has already been created.

Nice question, but I notice some reductionist bias behind it supposing that if we know the parts (interactions between taxa) we could predict the whole (functioning). Complex systems (such as benthic communities) just not behave in this way. Even implementing complex and well data feed quantitative models we just can have some short-term predictive ability. However, processes are far from being lineal and the occurrence of surprises is the only law…

There are many publications in this topic in Russia during 50 years at least

I totally agree with Diego's comment, the predictions about the system must be done by considering its complexity as a whole. But even so, the knowledge of how the different parts of the whole work sepparately usually allows us to explain certain patterns or abrupt changes observed at community level.

I think your deiced to be relate your main about it.

But i have chosen 1.found the relation between the response variable ( taxa) and independent variable such as: grain size, gradient of substrate,depth.by some analyses: pca,cca,ca . when we will be sure about kind of data ( normality)

after analyses ,2.you find which factor is more important, then discuss more about it.3.also used diversity index: shanoon - specially in the condition there are different habitat.

Have you dug into the biodiversity and ecosystem function literature at all? There's a great review by Stachowicz et al. in AREES in 2007.

We need to know the interaction between taxa, but not all the taxa that are in a special ecosystems. If you know some of them that you initially had considered as representative for evaluate it, you can evaluate the benthic functioning. Off course it depends of your original objective. the problems of some of us , is that sometimes we want to know everything there are; fisrt is almost impossible to know everything and, second it is not neccesary the most of times. Off course if you have more information about these relatonships, it should be better, but you will never know the total of interactions, because there are general interactions, chemical interactions, molecular ones?. First define the logistic limits, then the ecological limits and later you define until which level do you want to get information and, I think that there are an ecological limits that you can choose to obtain your answers.

Dear all, the question arose during a discussion on biodiversity valuation and I thought it would be appropriate to post this particular issue (although generalised) in the marine biology topic. It was about the interaction between meiofauna and resting stages of plankton in benthic environments - meiofauna predation affects plankton diversity in the benthic realm. Without knowledge on the interaction and how it works, it was argued that a correct assessment of the system's function was very unlikely.

I fully agree with some of the comments here in the sense that the interactions between taxa alone are not sufficient to understand the complex whole. The question did not allude to such logic, it merely questions whether we NEED the insights into the interactions (aside all the other important things such as abiotic regulation and genetic resilience, to name but two, etc.) to be able to assess function correctly. I.e. are insights into biotic interactions necessary to valuate function? or are the outcomes we are able to observe sufficient.

I'm inclined to say that the biotic interactions matter very much; without insights into them the observed outcomes cannot be used to explain future ones (Here, the previously mentioned stochasticity appears, or at least what we perceive as stochasticity). Of course, in attempting to valuate function, one is bound to revert to restricted components of ecosystems... we are far from understanding all interactions.

Thanks for the reference to the Stachowicz paper, I shall look it up.

Best wishes,

Jeroen

Dear Jeroen,

Here there is a book that despite been published 10 years ago is a milestone on BEF issues:

Biodiversity and Ecosystem Functioning; Loreau, M., Naeem, S., Inchausti, P., Eds.; Oxford University Press: Oxford, UK, 2002.

Cheers,

Carlos

Jeroen, thanks a lot for your question! So interesting!! I'm personallly interested in functional traits because they could provide crucial info about how the ecosystem works! But, with our current data only generic conclusiones could be stated. There is a long way to get accurate info about several scarcely-studied ecosystems, but we are on the wave!

Cheers

Rodrigo

Dear Jeroen,

Why do you refer to this concept as biodiversity valuation? What is there to valuate? Giving value to something means you have to explain the valuing context first. There is a normative context needed to understand what and how to valuate. Impacts of say dredging on benthos is not valuated by objective ecological parameters, but by valuing what kind of change in these parameters we (as a society, or subgroup) find deleterious and why.

Next, as to the relation between structure and functioning in (marine) ecosystems, it is too general to relate these two without being more specific. Reading general books and papers on the subject won't help you any further. We don't need to understand the "whole system", what is that anyway? All what has been said above is true, but too generic in nature. You need to specifiy the boundary conditions for the existence of the phytoplankton species, their developmental stages and their traits. Why the boundary conditions? Because it makes no sense quantifying the relationships too much; they vary in nature and extent too much for finding one solution to the equation... Understanding the influence of meiofauna on the resting stages of phytoplankton (diatoms?) may not be important, only if it forms a limitation to enumeration of a species. So, one has to ask the question: what determines the presence, biomass and numbers of the phytoplankton species, figure out its (or their) cycle(s) and estimate or assess the role the resting stages have in this cycle. It may not be important at all. An analogy: for the eggs and larvae of fish, we know that only under a certain threshold, reproduction may become limiting for stock development. Fish need to overproduce to make sure that the number of larvae won't be limiting...otherwise, stochastic events may well exterminate a species. Fortunately, there are quite some feedback loops and buffers involved, otherwise any species would not survive long. Usually, the specialised species, or by habitat, or by food, or any other niche description disappear first when under pressure.

So, you have to set up a proper schematic for the analytical dose-response relationships of the phytoplankton species involved. With this set up first, you can much better focus your questions and literature research. By the way, maybe the resting stages of phytoplankton could be of paramount importance for the fitness of meiofauna instead of the other way around?

Good luck with your quest, it is interesting indeed!

Arjen Boon

Dear Arjen,

Thanks for your interesting comment and insights. Below are some of my thoughts, probably oversimplified, as that is very easy to do in this context.

I agree that to solve the specific questions, say the effect of meiofauna on planktonic stages amongst many others, the hypothesis should be narrowed and the boundaries determines so to make it realistically testable with relevant outcomes for wider community understanding. A proper schematic conceptual interaction model with appropriate boundaries and parameterizations seems imperative. This will also allow integrating it into a more holistic modelling approach. However, that does not mean one cannot generalise and put the different questions into one integrative context. The question of do we need interaction knowledge to paint the complex picture of dynamic ecosystems, their biodiversity and how it behaves under various stressors (leading to the valuation step), seems answered by the reasoning in your comment... yes, we need this information if we are to understand and predict the outcomes. The paper by Stachowicz et al. 2007 in AREES, confirms the complexity of trophic interactions and is a good example of how to render specific, applied results into a more universally applicable scheme. I feel, however, that validation of such a scheme across different systems and their compartments is warranted... and should be guided by increased understanding of how complex interactions work. Suggestions to focus on critical species to come to a ‘quicker ‘understanding of ecological consequences on ecosystems may improve scientific focus (e.g. Russel et al. 2011 (Biology Letters), but in such an effort we will probably overlook important underlying dynamics, impairing our understanding in the long run.

I like your inverse reasoning, i.e why is phytoplankton not determinative for the meiofauna. It also renders the formulation of very specific hypotheses in this context perhaps unnecessary... since testing specific hypotheses will only serve to falsify the proposed hypothesis, not prove its universality in marine ecosystems (in one of the discussions, links below, it is argued that this is not feasible anyway). In fact, the inverse plankton-meiofauna interaction is probably true, as several works give an indication of the importance of meiofauna food resources to their structure and function. We are talking about very dynamic systems indeed, and general and specific understanding is probably more important than seeking to validate one-way interactions.

About the context of valuation, the question arose from an interesting discussion on valuation of biodiversity (aside the fact that it is heavily debated subject in literature) https://www.researchgate.net/post/Whats_about_the_value_of_each_species_of_natural_resources_Biodiversity_valuation_is_really_important_to_be_accessed

And

https://www.researchgate.net/post/Do_you_think_the_new_international_biodiversity_science_body_IPBES_will_improve_biodiversity_conservation_from_your_national_regional_perspective

Digging into the valuation debate doesn’t make it easier, valuation can take very different forms depending on the target audience, and the realisation that our current valuation methods are generally catered for a policy audience which need to be made aware about the non-monetary values of natural heritage. Awareness of biodiversity values beyond the scope of monetary terms is an important endeavour to be realised.

I can’t help but feel that there is an existent divide to some extent between understanding the systems (and their components) and using that knowledge in the valuation process to further conservation and sustainable management (I probably have been missing out on some of the literature). The exercise is not an easy one, and the danger of wanting to run before we can walk, or should we say, crawl, may undermine our efforts I think.

The quest continues, without a doubt... or should I say, many doubts!

Best wishes,

J

Dear all,

The discussions are very good. I think we must also consider the fact that ecosystems have certain things in common (on paper) but due to prevailing environmental conditions in various 'ecosystems' we may have different organisms 'functionally'. Genetic make up, adaptation and exposure to environmental conditions may render some organisms suitable to certain conditions as compared to others. Also, the ecological niche of similar organisms may differ even if they belong to the same trophic level. Nematodes for example have different genera that may function differently in the same ecosystem - even in the same micro habitat.

I am enjoying./ Could this also be a method of teaching biology? shoen!

Ecosystems are better studied in a step-wise or sectorial approach before pieces are now put together to get the big picture. For instant various likely food chains are considered before a food-web is constructed.

Thus a valuation of benthic organisms based on complex interactions between taxa is dependent on information or studies on the local flora and fauna constituting the benthic community.

I completely agree with Diego that even if we knew all the elements (species), the interactions between taxa and with other communities, and the external forcing from environmental factors we could not yet be capable to reasonably predict the whole (functioning), and this because of the complexity and intrinsic uncertainty of bio-ecosystems. Humans trying to understand the current state or predict the future condition of complex systems (such as benthic communities) regularly resort to simple, easily interpreted surrogates as parts of the whole complexity (e.g. biodiversity) that can be understood and even used by non-scientists to make planning and management decisions. Yet, the overall information we can gain from a set of indicators either structural and/or functional will never match that of the whole system, since each individual indicator carries only partial information. Thus, the set of indicators needs to be constantly re-evaluated and re-interpreted in the light of the increasing understanding of the whole organization and functioning of systems. Fortunately, the complexity of living systems emerges not from a random association of a large number of interacting factors but rather from a smaller number of key-controlling processes (Holling 2001; Gunderson and Holling 2002). Thus, much of the fundamental nature of systems can often be captured and described by single key-state variables, as many features of the system’s state tend to shift in concert with them. However, the problem is that each complex system is rather unique with non linear behavour and sudden shifts and surprises, and not always and anywhere the same key-state-variables are in operation and in the same way. Yet the challenge is to identify such single key-state variables, as representative of the many features of the system’s state that tend to shift in concert with them.

Dear Ingels

We have no expertise on the taxonomy of all the groups (especially the minor ones) living in the sediment interstitial. We do not have any knowledge regarding the biology and ecology of those unidentified taxa. Therefore it is impossible to accurately evaluate their functioning in the ecosystem. However, in terms of production, we can measure benthic production and can use it as an index of benthic ecosystem functioning.

What you think?

Best

Dola

Dear Dola, I agree that most of the biology and ecology of the "unknown" taxa hampers further developments, but efforts are currently being done to increase understanding of compartments of the benthic ecosystem in ways that surpass previous efforts, both in scale and complexity. Limited aut- and syn-ecological information can be used to get an initial picture of the observed interactions (whether they are real is something that needs consolidation with empirical data) that exist, and can give insights that may improve our understanding drastically.... perhaps such efforts can convince people to reinvigorate investment in the studies that investigate the behaviour and ecology of lesser- or unknown benthic species. Existing ecosystem models out there have the ability to predict certain dynamics but are currently vastly underused because of the lack of appropriate data but also a lack of benthic ecologists sitting down with modellers (and vice versa?)... synergy between these two sciences can improve our understanding, even if it was merely to understand the emerging properties of the models that are being run. We don't need to know everything to gain insights into what happens and what will happen, but we need to move on; using simple and increasingly complex models can shine the proverbial torch in the dark corridor of benthic ecosystem complexity. Production is definitely one way to go, but suffers the same problems in terms of what is producing what? One can give production estimates of the benthos, perhaps even report production of benthic compartments separately in a more general way (i.e. ignoring internal compartment dynamics) or taxon-specific groups... but surely we need to improve our knowledge on the complexity of interactions. One can continue in generality or one can dig deeper and unravel natural complexity... both avenues are important in my view. Best wishes, Jeroen

..in other words you work with the 'known' to get the 'unknown'. A continuous discernment of present information with increasing accuracy will reveal other information previously unknown.

Nice statement. I would also say working with the observable but not necessarily known is important. Of course correlation does not automatically mean cause-effect, but it may do. More knowledge will lead to much more knowledge. Best wishes J

Thanks for the complements Jeroen. Indeed correlation may not necessarily mean or interpret as cause-effect scenario but with a sound theoretical background, possible cause-effect relationships can be highlighted particularly with a statistical understanding of co-varying factors.

Cheers. Az

Dear All,

I really agree to what many of the members of this forum have said but I wish to add that while debating on this matter you should not forget the roles played by synergy, the obtaining environment and different components' interactions. All these seem to be interwoven into complex relationships that mankind has yet to thoroughly understand to become aware of how benthic communities clearly function.

Hi Jeroen,

Some rambling thoughts...

1. Is the question valid? Whilst 'functioning' or 'function'' is clearly an emergent property of the benthic assemblage there is no clear, set ,mechanistic relationship that can be quantified, simply because any given function is likely have show a unique response (which is why these studies demand a multi-faceted approach to evaluation of function - like Lars Gamfeld suggests). Any given measured function (take Mark Emerson and Dave Raffaelli's use of ammonia flux as an example) will be affected by key members of the benthic assemblage in a certain way, other measures will be affected differently, TOC flux, ammonia flux, mineral flux...all subtly different and affected differently. So the question re-phrases as 'what do we mean by function? as others have suggested, interactions and function are a bit like apples and pears in this instance.

2. Circularity? Surely a primary ''function' of the assemblage is best measured as...maintenance of the assemblage. So you get into the debate about assemblage dynamics and colinearity with diversity etc, plus there are statistical/sampling confounds (a la Huston's 'hidden treatments' critique of the BDEF debate) - the more interactions you study the more likely you are to encounter those having significant effects on whatever functions you measure, so the link between assemblage and function becomes probabilistically emergent - or is that the whole point??

3. Perhaps what is needed is explicit linkages between trophic network parameters and composite measures of function, or a new, less confounded (but probably infinitely more boring and less world-relevant) and more tractable, 'parameterisable' question... or a meta analysis of some kind....

Cheers

Andy

I like the way your response creates a philosophical angle to the discussion... and points out the fact that this is definitely no easy matter. Great to see the discussion growing further... nice platform to exchange ideas. The original question was of course designed to be thought provoking, glad it did. Have some thoughts...

1. The use of function was indeed holistically meant in that it consists of many emergent properties of community dynamics and processes... and hence there is indeed no set mechanism to be quantified to come to one magic number. It is not unreasonable to think however that despite the complexity, there are a finite number of processes and entities that contribute to functioning and functions. We're nowhere near identifying them all, but if we could, and if we could measure them all, could we come up with a magic number (or many numbers, although I doubt 42 will be one of them) to assess functioning or functions of our interest. All the different functions would indeed be affected slightly differently depending on (amongst others)......interactions between organisms and their environment in a vastly complex marine environment. Apples and pears... perhaps more like traffic jams and road networks (need to think about which one is function and which one is interaction)?

2. Circularity... great reasoning. Circularity in the fact that processes we measure affect organisms that affect processes that affect organisms that affect.......Depends on how functions are measured maybe (e.g. diversity is a proxy of functions, processes and interactions and easy to measure, but it's link with functions, processes and interactions remain elusive). Maintenance of assemblages is a tricky function(s) to measure (is it a proxy affected by the circularity you mentioned? it probably is). When I thing of function, I think of processes, interactions and organisms that affect processes, and the outcomes of those. I would certainly hope that the more we look at assemblages and function, the more significant relationships will become apparent. I guess it is the whole point, which is inherent to the question, can we identify all the different functions without understanding all interactions? The probability argument suggests that the more interactions we understand, the more significant our perception of functions will be... but can we jump the probability issue? Sometimes I think we should get rid of statistics in certain cases, a statistical p-level doesn't necessarily make ecological/biological sense or bare relevance to reality.

3. Brilliant suggestions... Maybe in the future we'll have an incredible complex network of parameterisable relations that reveal tons of functions for the benthic environment (linked to all adjacent realms of course), where we can chose and quantify those that are relevant for whatever we are studying or want to give advise on. Current modelling is restricted by knowledge.... hampered by the "parsimony choice" as a result of complexity we can't deal with yet.

I remain in thought... quite deeply after your insightful rambling thoughts

Cheers,

Jeroen

Hi Jeroen, are you familiar with the work of Vladimir Kostylev? Scope for Growth maps produced by Kostylev and Hannah (2007) are based on ecological theory that relates species life history traits to the properties of the environment. Characteristics of habitats impose selective forces through a variety of biotic and abiotic factors; these factors affect the fitness of individual organisms by modifying their growth rate, survival, fecundity, etc. The two major selective forces are physical disturbance and adversity of the environment. Large scale functional maps showing the distribution of habitats where organisms with particular life history traits are likely to flourish following r-K theory. Kostylev, V.E., and Hannah, C.G. 2007. Process-driven characterization and mapping of seabed habitats, in Todd, B.J., and Greene, H.G., eds. Mapping the Seafloor for Habitat Characterization: Geological Association of Canada, Special Paper 47, p. 171-184.

Hi Ellen, thanks for that... great paper.

Hi Jeroen,

This is a very interesting question, admittedly I haven't read through everyone's response so hope this adds something new. One angle to take is the ecological limiting factors. A key liming factor is likely to affect the maximum values (and associated changes in the variability) of the ecosystem function response variable. This makes sense given that 1) organisms are patchily distributed over space and time and 2) ecosystem function is the product of multiple and often interacting factors, not all of which are possible to measure. Inspecting x-y scatterplots of data clouds can sometimes reveal trends detectable at the upper limits of the distribution, known as factor ceiling distribution (see Thomson et al. 1996). Factor ceiling distributions are becoming trendy and are useful to assess the overall/net affect of changing biological or environmental factors to ecosystem function, which is handy given the complexity of interactions that constitute them! Also, check out the paper by Simon Thrush.

Thomson JD, Weiblen G, Thomson BA, Alfaro S, Legendre P. 1996. Untangling multiple factors in 7 spatial distributions: lilies, gophers, and rocks. Ecology 77:1698-1715.

S. F. Thrush, J. E. Hewitt, and A. M. Lohrer 2012. Interaction networks in coastal soft-sediments highlight the potential for change in ecological resilience. Ecological Applications 22:1213–1223. http://dx.doi.org/10.1890/11-1403.1