The niche is a basic concept in ecology, and seems like something that everyone should be able to easily define – and to define consistently. Yet, recent conversations have made me realize that there’s a lot more variation than I would have expected in how the niche – especially the realized niche – is defined.
The crux of the debate seems to come down to whether people say the realized niche is limited by just competition, or whether they say it’s influenced by all interspecific interactions. There’s also a debate on whether dispersal plays a role. Once I realized this and started talking to others about it, there were two main kinds of reactions:
1) “OMG, yes, isn’t it weird that some people define it that other way? Clearly they are wrong.”
2) “Wait, what? Not everyone agrees that X is what determines the realized niche?”
To back up: The original Hutchinson “Concluding remarks” paper focuses on competition as the factor that limits organisms’ distribution; that is, in that paper, the realized niche is smaller than the fundamental niche due to competition. Since then, though, there has been a lot of work updating the niche concept. This includes relatively recent work, and work that adds in mutualism, which can make it so that the realized niche is larger than the fundamental niche.
When I teach about niches, I teach that the fundamental niche is where an organism can occur based on abiotic conditions, and the realized niche is where it actually does occur, given the presence of interacting species. I tend to gloss over the effect of mutualists on the niche in my Intro Bio course, but I used to focus on that more when I taught Ecology. This mostly matches what the textbook we use (Morris et al’s How Life Works) says:
The fundamental niche comprises the full range of climate conditions and food resources that permits the individuals in a species to live. In nature, however, many species do not occupy all the habitats permitted by their anatomy and physiology. That is because other species compete for available resources, prey on the organisms in question, or influence their growth and reproduction, reducing the range actually occupied. This actual range of habitats occupied by a species is called its realized niche (Fig. 47.2).
The main thing I think is left out of that definition is the possibility for facilitation to make the realized niche larger than the fundamental niche, but, again, that is not a topic I tend to emphasize at the freshman level. The way I teach about the niche also matches what is in Begon et al., which is my go-to reference for checking on this sort of thing:
“Usually, a species has a larger ecological niche in the absence of competitors and predators than it has in their presence. In other words, there are certain combinations of conditions and resources that can allow a species to maintain a viable population. This led Hutchinson to distinguish between the fundamental and realized niche. The former describes the overall potentialities of a species; the latter describes the more limited spectrum of conditions and resources that allow it to persist, even in the presence of competitors and predators.” (page 31 of the 4th edition)
So, given all that, I was surprised to learn that others (including some of my colleagues here at Michigan) hold the view that the realized niche is only influenced by competition. In recent conversations I’ve had with others about this, several people at other institutions said they shared my view but also have colleagues who hold the view that the realized niche is only influenced by competition. This has me wondering what the split is in terms of how many people hold the different views. Hence this post.
I can see that it might be possible that people might think one thing but teach another in the interest of trying to keep things simpler. So, I’m also wondering:
At the risk of biasing the poll, I will say that I find the view that only competition restricts the realized niche to be surprising. What, then, explains why really large bodied Daphnia don’t co-occur with fish? It’s not that they’re not good competitors.
Finally, another issue that has come up during these discussions is whether the realized niche is influenced by dispersal. Let’s consider this clicker question that I’ve used in class:
60 common starlings were released in Central Park in NYC in 1890 by someone trying to introduce all the birds mentioned in Shakespeare’s plays to the New World. There are now millions of starlings in the US, and they are common in Ann Arbor. Prior to 1890, Ann Arbor was:
A) part of the realized niche of starlings but not their fundamental niche.
B) part of the fundamental niche of starlings but not their realized niche.
C) part of the fundamental and realized niches of starlings.
D) not part of either the fundamental or realized niche of starlings.
I give the correct answer as B. But some colleagues of mine argue that dispersal does not influence the realized niche. Everyone is in agreement that Ann Arbor was within the fundamental niche of starlings prior to 1890. The question is whether it’s okay to say that the realized niche of starlings was extended by them being introduced into Central Park. I say yes, but others say no. What do you think?
I can understand how this is more of a gray area – if we are saying that the realized niche is influenced by interspecific interactions, then it’s not clear what to do about the influence of dispersal. But, then again, how else would you characterize the change in distribution of starlings? I’d love to hear people’s perspectives in the comments.
All of this is reminding me again of one of the things that I really like about teaching – it forces me to think harder about things that I thought I knew. I will be interested in hearing more about what others think about the realized niche, and how they teach about it!
Dynamic Ecology 
I’ve felt similarly awkward teaching this concept. I think the problem is that the only logically consistent definition includes all interspecific interactions plus dispersal; otherwise, we have the problem that we would have to declare that some species is living (with r>0) in a place that is not part of its realized niche, and that just doesn’t make any sense (and yes, I know this puts me in the “(1) OMG…wrong” camp. BUT – this definition is logically consistent but not much use, because it’s just the same thing as the geographic range. It would be more useful, but much harder, to define a series of fractionally different “niches” from the abiotic niche to the abiotic+competition niche to the abiotic+competition+predation niche, etc… that is, to partition occurrence among processes. But good luck covering that (and how to do it) in half a lecture in Intro Ecology 🙂
Half a lecture?! We don’t have that sort of time in Intro bio! 😉
RE starlings in Ann Arbor, I think it helps to keep niche space and physical space separate, as Hutch did in Concluding Remarks (remember all that B business?). To say that “Ann Arbor was part of the realized niche”, under this framework, doesn’t even really make sense. Instead, I think I would choose an answer that read, “The starlings realized niche existed around Ann Arbor”, or something similar. It’s my opinion that including dispersal as part of the definition of a realized niche conflates niche space with physical space, and so should not be a part of the story. All in favour? All opposed?
Yeah, I remember back when I was TAing intro ecology for David Ehrenfeld at Rutgers, he was very keen to make sure students understood that “the niche is not a place”.
I agree that it’s important to be clear that niche space and physical space are completely different kinds of spaces. However, it seems like the distinction can get fuzzier in cases where spatial processes are affecting a given niche. Say an organism has a niche that relies on edge effects – it lives at the interfaces between two ecosystems. Its niche (both fundamental and realized) is still a conceptual set of properties that describe locations in which it can live, but one of those properties does actually have to do with physical space. I would argue that in that case it’s reasonable as a shorthand to talk about physical space that is part of a niche, because that’s actually one of the constraints defining the niche. Thoughts?
Tim and I continued this conversation over e-mail (his browser was having problems replying here), and he raised some really interesting points. Here’s a summary:
First, Tim pointed out that the environmental conditions at the border between two ecosystems are a set of environmental conditions just like any other, and can define a niche. In response, I drew a distinction between species that live specifically in the conditions at the interface between two ecosystems (e.g. tide pool creatures living at the intersection of land and ocean) and species that live at the interface because they need access to both ecosystems (e.g. most pinnipeds). I agreed that in the former case you can define a niche based on the conditions at the interface, but argued that the former case seems to be harder to define as a niche without invoking physical space somehow.
Tim had the excellent idea of fixing this problem by changing the grain at which you are measuring the conditions and just specifying that the environmental conditions associated with both ecosystems must be present within this larger area. You could argue that specifying a grain is sort of invoking physical space, but I think this idea does get at the idea of defining a niche that is dependent on physical space without directly invoking physical space. However, there can be more complex spatial effects that I’m not sure this fixes. I illustrated this with a toy example of a hypothetical species that (for some unspecified reason – perhaps a developmental constraint) needs to travel from environment A, to environment B, to environment C (without ever going to an environment in the wrong order). You could have a region of space in which this is not possible, despite all three environments being present.
Tim rightly observed that an environment could also fulfill the A -> B -> C requirement through temporal variation, so you can’t define the niche of this hypothetical creature exclusively in terms of space and more conventional variables used in defining niches. Based on this, he suggested that more conventional niche variables are in some way more fundamental. I definitely see the appeal of leaving out the messiness of spatiotemporal variation as a simplifying assumption, and I think it’s appropriate in most cases. I just think it’s important to define niches in a way that allows us to talk about the creation of new niches based on spatiotemporal relationships between ecosystems when it’s relevant.
Ultimately these are pretty subtle distinctions (particularly relative to the question of how to define realized niches), and neither of us feel that strongly, but it was still interesting to discuss!
Completely with you. Realized niche is the net result of biotic and abiotic constraints. Whether you can reach it or not (through active or passive dispersal) is a different thing.
Note that ‘reach’ is used figuratively.
Yes, I wonder if focusing more on range (which I currently don’t spend a lot of time talking about) would be useful. More to ponder!
Especially for intro bio, I’ve had the same thought.
Interesting question – I am surprised to hear that some people insist realized niche is only about competition. (or negative interactions) in 2015.
To me the starling question is sort of unanswerable. New continents are just sort of a whole different game/process. And in general I conceive of dispersal as distinct from realized niche processes. The 3 filter (abiotic, biotic and dispersal) framing feels like a more up to date description of niches (I see Jan has already tweeted the Soberon paper on this framing). Its not just new continents either. There are a lot of papers showing plants in Europe are still not in equilibrium after the last ice age and still haven’t physically occupied much of their “potential realized” niche yet due to dispersal limitation.
You ommitted one case in the literature. That is where the realized niche extends just a bit beyond the fundamental niche due to dispersal from sources inside the range to sinks just outside the range (r<0 there so not in the fundamental niche, but species is found there so in the realized niche). Pulliam talks about this case.
Your point about sinks is an excellent one! That would be an interesting one to cover in an ecology course, but is definitely beyond where we go in Intro Bio.
Via twitter, Jan Engler suggested this Soberon paper: http://onlinelibrary.wiley.com/doi/10.1111/j.1461-0248.2007.01107.x/abstract
and said:
the main difference is whether you see niches in environmental or geographic space. Migration affects the latter.
As an evolutionary biologist, I don’t deal with the concepts of niche and habitat daily, and I wasn’t aware of this controversy over realized niche. I have a couple simple questions. Is Ann Arbor a habitat for starlings, or a location with starling habitat? Ditto, Central Park? Is a niche a set of ecological conditions, or a place which can be described in terms of ecological conditions?
I would have thought that the arrival of starlings in Ann Arbor (a sad day for local cavity nesters) had no effect on the niche of starlings, as long as they did what they always did. I can see that their arrival could change the realized niche of hole nesting bird X with which they compete, however, if we imagine that starlings out-compete species X for certain types of holes. But am I wrong in the way I think about habitat and niche, as being defined by biology and ecological conditions and not by addresses?
aargh. I took a break while writing the above, and see that others are saying the same thing…
Possibly nit-picky aside: For anyone who teaches the idea that facilitation can extend the realized niche, even beyond the fundamental niche, are you *sure* you know *exactly* when that can happen? The scattered verbal literature on the effects of facilitation on species’ realized niches and/or geographic ranges appears to contain some serious mistakes (or at least serious unrecognized ambiguity), and unfortunately it’s not a topic that’s been treated by a large body of mathematical theory yet.
I think the perception of how much of realized niche is influenced by competition vs. other interaction types is strongly influenced of your (or rather your profs’!) organismal background. from my student time for instance I would infer that a plant ecology professor has a different take on that than an animal ecology professor.
My view is that the fundamental niche should be defined simply as everywhere the abiotic environment would allow for a species to exist at reasonable densities (re: densities that allow for a viable population). And the realized niche is the set of climatic variables everywhere the species actually occurs *for any reason*.
I’m not sure what is gained by saying that a particular process that expands the range of climatic variables a species actually lives in “doesn’t count”. Especially since it is almost impossible to completely disentangle pure dispersal from competition. If you imagine two valleys in the Appalachian mountains with different salamander assemblages (and slightly different temperatures/precipitation/etc), a particular salamander may not be able to invade the next valley over because 1) it’s truly unable to reach that valley (pure dispersal limitation), 2) it can easily reach the valley but is outcompeted there (pure competition), 3) it can reach the valley, but not in large enough numbers to establish a viable population because immigrants are killed by interference competition/whatever before a viable population can exist (competition+dispersal), or 4) it can reach the valley and isn’t stopped by competition, but cannot live in that climatic regime (outside the fundamental niche).
Really, the starlings might be an example of facilitation. Humans facilitated their dispersal, and heavily altered the competitive landscape such that starlings were able to successfully invade. It’d be really great if we had a time machine and could release 60 starlings in New York before Europeans colonized it (or even before Native Americans arrived) and see if they were still able to successfully invade the rest of the continent.
Folks above correctly point out that the realized niche is not a place, but I have a hard time believing that everywhere starlings currently live in North America falls into the same abiotic envelope as everywhere they live in the Old World…
That’s a really interesting point about the possible role of humans in the success of starlings once they were introduced! I hadn’t considered that, but agree that it’s quite possible.
Very interesting discussion! Yes, I had the same issues when I started teaching Ecology to undergrads (and keep wondering every year since…). When we discuss realized niche I only gave the “easy” competition-predation-parasitism examples (mainly referring to competition) because they are easier for the students to grasp. I would agree with the “for any reason” notion of realized niche by fieldofbullets. Regarding that, I think realized niche is not a place as I perceive it, but perhaps a place-instance-trait combination (inserting the time dimension and the notion of function) or a place-instance occurrence (taking into account only the time dimension)?
It seems to me that the logical conclusion of assuming that facilitation cannot influence the realized niche is that obligate mutualists have no realized niche!
For my undergraduate research project (in Ann Arbor!), I explored the trait dimension of the niche. I mapped the leaf traits for a forest community to see how much the competing species traits overlapped and if there were holes in the trait space of the community. I found an invasive shrub greatly expanded trait space for the community, possibly occupying an empty ‘niche’, or at least a new place on the leaf economic spectrum.
As for the fundamental vs. realized niche question, I’ll echo what other folks have said about being very clear what niche space you are in (environmental, geographic, trait?, …). The answer to starling question certainly depends on it.
I’ve not had to teach about the niche yet, but I’m very glad I’ve been given the chance to think about it and it’s ambiguities again. The discussion reminds me of the many reasons I chose to move my later research away from the niche concept. Thanks Meghan for the provocative post!
Just to add fuel to the fire, if we include all interspecific interactions as contributing to a species realized niche, what constitutes the fundamental niche of a predator, herbivore, or parasite? Is there any use in distinguishing between the two? And is Hutchinson’s definition of the fundamental niche fundamentally inappropriate for parasites that are transmitted directly between hosts?
I was thinking something similar. Is fundamental niche something quantifiable? If all species interactions influence niche, eventually every species is unable to survive when we eliminate interacting species one by one (I don’t know much about autotrophs). Suppose there is a specialist consumer that cannot survive without the specific resource. For this species, if we try to quantify the temperature axis of the fundamental niche (in the absence of the essential resource), under any temperature, it cannot survive. One may say that because temperature is not the cause of death, and thus the temperature niche (at fundamental niche level) still exists. But I feel strange about it.
This is one good reason (among others) why we should stop teaching Hutchinson’s notion of the niche as an n-dimensional hypervolume. In the humble opinion of someone who’s never taught intro ecology or intro bio himself. 🙂
Toward that last point about starlings, it seems like you have to draw a line somewhere. Perhaps some predatory species could occupy a space if it’s prey species were present. If the space is part of the prey’s fundamental niche, but not it’s realized niche, based on dispersal or other limitation, could we not then argue that the site is part of the predators fundamental niche? If we extend the logic beyond this it seems like one loses all meaning in the concept all together.
I prefer the Chase and Leibold definition of niche as the environment an organism needs to maintain a viable population and the effects that it has on its environment. Leibold first started fleshing out these ideas while we were junior professors teaching ecology and finding it very difficult to avoid confusing students with the niche concept. Thus, I agree with Jeremy that , while initially appealing, the ideas of fundamental and realized niche are on closer inspection fatally flawed and should be abandoned.
It is too bad Mathew hasn’t gotten more traction with that way of thinking about niches, I for one think it’s a big improvement over Hutchinson’s way of thinking of niches.
Although if I had to pick one way of thinking about niches, I’d probably pick Peter Chesson’s. And yes, I do think you could teach that to intro ecology undergrads (in a here’s-the-gist way, obviously, just as with all the other ideas we teach undergrads). But easy for me to talk, since I’ve never had to do it! (Well, I’ve taught it to advanced undergrads in a population ecology class, but that’s not really the same.)
In terms of definition, I also want to make sure whether ‘natural selection’ specifically assumes evolution. If it assumes evolution, why do we often say ‘evolution by natural selection’?
http://sandwalk.blogspot.tw/2013/11/natural-selection-may-not-lead-to.html
A bit off topic for this post, but interesting and super important.
It’s well accepted that evolution and natural selection are not the same thing, and that selection can occur without causing evolution if variation is not heritable. As soon as heritability is greater than 0, evolution will occur (I’m assuming only genetically based inheritance here). So that’s one reason for “evolution by natural selection”. Evolution is not assumed.
Another reason is that evolution can occur by forces other than natural selection. Gene flow, mutation, and genetic drift can also cause evolution.
Yes, it is not about niche, but it is also a definition problem. Even for fundamental/realized niche, from this blog I only learned people use different definitions and am not sure what the true definition may be. This is an important problem for ones teaching the subject. When I notice that there are variable (and contradicting) definitions even for a fundamental concept, I usually tell students different definitions exist. I also tell them I do not know what the true definition is, but the important thing (within the course) is to understand the concepts rather than focusing on how we call them. But this is not responsible of me, and students would not know what to do when they are taking standardized exams outside the course. And even for natural selection, I am just hearing different opinions and do not know how to reach the truth. We need a way to check if a given definition is correct when different books have contradicting definitions.
Pingback: How do you define omnivory? And what do you teach about trophic levels? | Dynamic Ecology
Late to the party, but still a few comments
* Generally, I too tend to lump everything in the realized niche (all biotic + dispersal)
* However, this means that the current distribution in environmental space, i.e. what we fit in SDMs is the realized niche – there has been plenty of discussions on this in the literature that I don’t want to iterate here, but let’s say some people have reservations regarding this definition
* Dispersal can distort the fundamental niche not only through disequilbrium (not there yet), but also through source-sink dynamics, which allows for an expansion beyond the fundamental niche. Pagel and Schurr propose a way to separate such differences statistically, see Pagel, J. & Schurr, F. M. (2011) Forecasting species ranges by statistical estimation of ecological niches and spatial population dynamics. Global Ecol. Biogeogr.
* We had two workshop on dynamic distribution models a while ago in which the discussion about the different niche concepts surfaced repeatedly. In the end McInerny & Etienne wrote a series of three papers on the problem, the first one is this McInerny, G. J. & Etienne, R. S. (2012) Ditch the niche – is the niche a useful concept in ecology or species distribution modelling?. J. Biogeogr., 39, 2096-2102.
* From the same workshop, there is a proposal for a framework about a demographic definition of the fundamental niche that basically puts the idea of fundamental = growth in isolation, realized niche (all biotic + dispersal) in a more formal statistical framework Schurr et al. (2012) How to understand species’ niches and range dynamics: a demographic research agenda for biogeography. J. Biogeogr., 39, 2146–-2162.
* Finally, all the discussion above applies for Hutchinson’s framework in which the niche is spanned by the environment, not for alternative views where the niche is spanned by the functions of a species http://evol-eco.blogspot.de/2015/07/can-there-be-periodic-table-of-niches.html . I have to admit that ultimately, I tend to think of that as the more fundamental property, while Hutchinson’s “fundamental” niche rather follows from the traits / properties of a species.
There’s a lot here to unpack, but Ill be brief.
1) I’ve taught niche theory dozens of times and I don’t see it as being especially challenging to teach or grasp. Yes it should be taught. It’s obviously one of the most fundamental ideas in ecology. Kids learn it (or versions of it) multiple times in primary school. Most textbooks still have a very old view of the fundamental and realized niche. But it is pretty easy to talk about conceptual developments over the last two decades.
2) I’m surprised but delighted that 95% of our colleagues think facilitation influences the realized niche. When we proposed this idea ~15 years ago, it was seem by some as a bit radical and Science magazine labeled us “renegade ecologist”.
3) Jeremy, what is “verbal literature”?
4) Jeremy, what on earth does this mean? “are you *sure* you know *exactly* when that can happen?”
5) Jeremy, would you care to articular the “serious mistakes” you allude to? Because the link you provide doesn’t. It goes to a post about relative fitness, nurse plants, etc. that isn’t relevant to niche expansion.
6) Meg, there is a huge literature on this, eg. the paper that originally developed the idea that facilitation could influence the realized niche, etc:
Inclusion of facilitation into ecological theory. Bruno et al 2003: http://www.cell.com/trends/ecology-evolution/abstract/S0169-5347(02)00045-9
The growth of theory beyond this:
Facilitation and the niche: implications for coexistence, range shifts and ecosystem functioning. Bulleri et al 2015: http://onlinelibrary.wiley.com/doi/10.1111/1365-2435.12528/abstract
Lots of debates about and critiques of the concept:
Positive interactions in ecology: filling the fundamental niche
Rodriguez-Cabal et al: http://library.queensu.ca/ojs/index.php/IEE/article/view/4335
Niche expansion by positive interactions: realizing the fundamentals. A comment on Rodriguez-Cabal et al. Stachowicz: http://library.queensu.ca/ojs/index.php/IEE/article/view/4448
And many cool examples experimentally documenting geographic realized niche expansion via facilitation, e.g.;
Positive interactions expand habitat use and the realized niches of sympatric species
Crotty and Bertness: http://www.esajournals.org/doi/abs/10.1890/15-0240.1
Regarding point 6: That is the paper I link to in the post where it says “work that adds in mutualism” in the third paragraph. (I haven’t edited the post. It’s been there the whole time.) It was a really important update to the niche concept, IMO.
Regarding point 2: I agree that it’s really interesting that 95% of our respondents think facilitation influences the realized niche. It’s not that long ago that we were all being taught that the realized niche is always smaller than the fundamental niche.
Thanks Meghan. I was only trying to be helpful by pointing out a small sample of the relevant literature for Jeremy, readers, etc. I wasn’t criticizing you for not citing me. (Although technically, you didn’t, you linked to the paper, which isn’t the same as citing it).
just found a nice commentary piece in the published literature on exactly this topic of the post:
“Invasive species: reality or myth?”
by Shah and Shaanker
http://link.springer.com/article/10.1007/s10531-014-0673-y
focussing on camels rather than starlings but in a very similar context.
Thanks!
Pingback: Favorite good (or bad?) examples of “operationalizing” vague ecological concepts? | Dynamic Ecology