Survey: Life span of marine, ornamental fish in captivity

[billsreef]

First it will tell us the average lifespan of different species held in captivity by above average skilled aquarists (because that is what people who hang out at forums are).

A number of years ago I would have agreed with you on this. However, over the last 5 or so years, the demographic of the average forumite has changed with Newbies and folks that just want what they want, without regard to the needs of a given species, becoming more common.

What I am trying to say is that I want people to add fish that dies because of accidents, because of malfeasance, because of stupidity and neglect, because it will show up as hardiness in the results.

Considering that even the most hardy fish will easily and commonly die from such things, and even a number of so called sensitive fish will live under proper conditions and care, I don't see the relevance.

To have much meaning, I think you really need to collect the data on experience level of the aquarist with that longevity in captivity data, as well as separating out such things as death due to accidents or neglect.

 

[snorvich]

Very simply, since you are only capturing deaths, you are ignoring successes. So my G. bellus pair, which are going on six years cannot be tabulated because they did not die. Or the fish that dies within a week from poor shipping practice will also skew the data inappropriately. All in all, these factors aside, the fact that your sample is not random makes any real conclusion about the population of no value since there is sampling bias. In any case, the best way to find out whether a fish is hardy is simply to ask on this forum. You will find much expertise to assist. Now if you are trying to capture information to publish on your own website or board, the flaws of the survey will make that information of no real value.

 

[Anemone]

Both Billsreef and Snorvich make excellent points. The data you are collecting is close to useless. I could ask 100 newbie hobbyists how long a yellow tang will live, or I can ask 10 experienced hobbyists - the results will be vastly different. And with no attempt to control for that component (hobbyist experience), the data is next to useless. Combine that with the fact that someone who has a 10 year old fish still living can't contribute, while a person who has killed five of the same species in the past year can, makes your survey an anecdotal presentation of little real value.

Sorry,
Kevin

 

[Agathos]

Considering that even the most hardy fish will easily and commonly die from such things, and even a number of so called sensitive fish will live under proper conditions and care, I don't see the relevance.

Yes, total annihilation of fish population where even the most hardy fish die do take place. And this is one of the reasons why there will be a bias towards "quick deaths". (I am aware of this effect and have included it in the hypothetical graphs above.) But there will be crashes where the hardiest survive, or only the most sensitive die, and this differentiation in survival will be evident in the graphs. For instance, I suspect that clownfish will have relatively less die-off in the first years compared to more sensitive fish. And this comparative distribution of deaths is what I am looking for. A fish that has relatively less deaths in the first year compared to another fish, but they both are able to live to 15 years, is easier to care for then the other. And this is of value to people who want to know more before purchasing new fish.

To go in a little more depth. The distribution of death graphs as displayed above have two important features: How many years with observed deaths they span, and their shape. The former tells us how long the fish may survive in our tanks (again, this relies on people with lots of experience participating), and hence it says something about the longevity of the tank in captivity. It is enough to have one single fish die after 20 years to prove that this fish may live for 20 years if cared for properly (and bought when young enough). The latter (the shape of the graph) tells us when people have experienced deaths with this fish after introducing it to their tanks. The shape will start with a peak in all cases (because of people buying fish they are not able to care for properly with the result that the fish die within the first years), but this early die-off will be different for different species because of the fish' hardiness. Hence this is exactly what I am looking for.

Again, let me emphasize that this study is all about comparing different fishes. There will be complete die-offs happening in each year for each fish species, but this is statistical noise that may not be a problem if enough people participates since this noise is uniformly distributed across the study. There will also be a bias towards early die-offs which will happen to every species but not identically, thus giving us valuable data for comparing their hardiness. Another immediate conclusion from the study is how long we might expect our fish to live in our tanks if we care for it properly. Of course we might not know the age of the fish when we buy it, but from looking at the data we know how long it at least may be able to survive if it is young and if we care for it properly. Such data may even be an incentive to people for buying young fish, or even steering away from species that just do not have a long lifespan in captivity.

Again, if people respond to the study we end up with empirical data that may be used for qualified expectations on different species' comparatively hardiness and longevity in our tanks. I still believe that this would be of value to us, especially the inexperienced hobbyists. I therefore continue to urge people to participate. It takes only a few minutes .

And to comment on the fact that this study only focuses on fish that are dead and not those that still live. Yes, that's the point of the study. For each and every one of us (hopefully) this means that most of our fish will not be added to the study now...but perhaps next year ? I have no plans of discontinuing the study now, I believe it will grow and become more and more valuable as each year passes on and more and more data is included. And besides, even if most fish are still alive resulting in a skew towards early die-offs, this is again a universal attribute with the study that encompasses all species, and hence it will not reduce the possibility of comparing death occurrences of different species. It is just a matter of normalizing the data to immediately see differences between species. If I were to calculate average lifespan, or even average time of death, on the other hand, of fish species, it would of course not be possible with this study, but that is not the point (at least not any more).

I hope this cleared things up a bit.

 

[Agathos]

Very simply, since you are only capturing deaths, you are ignoring successes. So my G. bellus pair, which are going on six years cannot be tabulated because they did not die. Or the fish that dies within a week from poor shipping practice will also skew the data inappropriately. All in all, these factors aside, the fact that your sample is not random makes any real conclusion about the population of no value since there is sampling bias. In any case, the best way to find out whether a fish is hardy is simply to ask on this forum. You will find much expertise to assist. Now if you are trying to capture information to publish on your own website or board, the flaws of the survey will make that information of no real value.

I have commented on the first part of this in my previous post.

I believe it would be much better to consult the experience of (hopefully) thousands of hobbyists than to ask here and get the response of a few. Empirical evidence is always better than anecdotal evidence.

I have no intention of publishing the results anywhere besides the places where I ask people to contribute to the survey. That is per today here on Reef Central as well as on three Scandinavian forums.

When it comes to flaws in the survey I believe I have commented on those in my previous post.

 

[Agathos]

Both Billsreef and Snorvich make excellent points. The data you are collecting is close to useless. I could ask 100 newbie hobbyists how long a yellow tang will live, or I can ask 10 experienced hobbyists - the results will be vastly different. And with no attempt to control for that component (hobbyist experience), the data is next to useless. Combine that with the fact that someone who has a 10 year old fish still living can't contribute, while a person who has killed five of the same species in the past year can, makes your survey an anecdotal presentation of little real value.

I am not trying to find out "how long a yellow tang will live", I am trying to find out "when yellow tangs have died after being introduced into new tanks". This is two vastly different objectives.

 

[Agathos]

I wish I could modify my first point to make it a lot clearer what I hope to gain from this survey. I see know that I voiced myself very poorly and that this is one of the reasons why people believe I expect to get something out of the survey that I agree is impossible/very difficult. Oh well .

 

[kactusficus]

so, how is this survey going?
I don't know why I can't see the result page, I'm always redirected to the poll website's homepage.

I'd like to know more about your results, is there something you can tick in the options of your survey preference maybe?
I tried to reach the websites's online help, to know why I couldn't see the result page, but they didn't answer...

 

[Agathos]

so, how is this survey going?
I don't know why I can't see the result page, I'm always redirected to the poll website's homepage.

I'd like to know more about your results, is there something you can tick in the options of your survey preference maybe?
I tried to reach the websites's online help, to know why I couldn't see the result page, but they didn't answer...

Hi, sorry for not responding earlier. I have been really swamped and haven't checked in here for quite a while.

My idea is to make some graphs that displays the relative mortality between different fish species at different time points after introduction to a new tank. In other words: Graphs where it is possible to see which species typically live long and which typically live short, and when they can be expected to die after introduction to a new tank.

The study will not say much about their biological lifespan (because we rarely know how old the fisk are at he time of acquisition), but it will say something about their relative sturdiness after moving and how long the maximally can be expected to live. This require that many people respond, otherwise sampling errors will make conclusions very hard.

Below I have attached some real results from the study for the three species I have most data. It's not enough data to at all make any strong conclusions but I will use this graph to explain a little more about what I expect to get from the study.

The X-axis show the years after introduction when death has been reported for each species. The curves for each species stop when no deaths have been reported. This immediately tells us something about how long we can at maximum expect out fish to live after introduction to our tanks, and this s related to their biologic lifespans and to a certain degree their sturdiness. If we look on the graph below we see hat respondents have reported that A. ocellaris may live to at least 15 years after introduction, while S. vulpinus has not been reported past 9 years, and A. flavissimus not past 8 years (after introduction). Of course I expect this to be adjusted as more people respond.

In addition the graph tells us something about when the fish typically die. For each year (on the X-axis) higher mortality than the average for all three species, give a positive value, while lower mortality than the average give a negative value. It's way to early to draw any real conclusions here, but if we look at the time 6-9 years it seems like A. ocellaris has lower mortality than the two other species. This is probably caused by A. ocellaris's long lifespan, and if you have succeeded at keeping it alive tis log you are probably an experienced and skilled hobbyist that will keep it alive for more years. For the other two species it looks like we are approaching some kind of limit at this time.

Additionally, we can look at the time 0-1 years. Not much data is available, but if we should jump to a conclusion it looks like A. flavissimus survive this period a little better than the other two species. A probably explanation may be the A. ocellaris is a fish that every newbie bys and hence it suffers from lots of deaths caused by beginners' mistakes.

Thanks to everyone that respond to the survey!!

 

[DaveP]

I think this is a great idea. However, I can't list my fish that I still have because they are alive.

Also - it would be interesting to correlate tank size to this study (not just for tangs )

 

[snorvich]

I think this is a great idea. However, I can't list my fish that I still have because they are alive.

Also - it would be interesting to correlate tank size to this study (not just for tangs )

Actually, that is just one of many "nice to have" correlations. If one wants to run a multiple regression, one needs a whole lot more. As it is, there is really no value to the survey.

 

[Agathos]

As it is, there is really no value to the survey.

Of course there is. Already from the results presented above it is possible to gleam conclusions of value. They may not be of value to you, but that is something entirely different.

I could have expanded the survey to include other parameters like age of hobbyist, size of tank, filtration method, etc, but I never intended to make those correlations. That was not the purpose of the study. This study is primarily meant to investigate interspecies differences when it comes to mortality - not why they actually die. Of course getting the reasons would be very interesting too, but it's too late for that now.