Couple of Questions

[Experimenter]

Dr. Ron,

First, Great article! This is the kind of research I wish we could do more of in this hobby. Too bad it is so expensive to do. Also, I apologize for the length of this response. As a reviewer I sometimes get carried away!

Second, I am wondering about your assumption of normality of some of the variables. I am most interested in the tank factors, since honestly I don't really understand the chemistry or much of the biology of what is going on. In particular, I think it is probably an error to assume that (1) tank age and (2) tank size have normal distributions. Tank age, by definition, does not (it should be positively skewed). Since tank age is probably one of the most interesting factors (at least for me) I see this as a potential problem. I wonder what the correlations would look like if you performed a log transformation on tank age and tank size? It may not change anything, but you never know.

Third, I am not sure how you measured "water changes." Is it simply "yes I do them" or "no I don't"? It would appear to me you could calculate a better measure of how much water is exchanged. Maybe such a measure would be the percentage of tank volume exchanged in an average one month period. This way you could recode the people who change 10% 4x/month to a measure of the monthly exchange (you could also do this for people who do it less often). I know this would be imperfect, but I think this issue is more important than your results show - and this could be a function of the measure and not the real factor.

Fourth, I am getting the picture from your results that tank age plays an important role in the build-up of nutrients and other elements. The problem is, we cannot do anything about the age of our tanks, per se (without tearing them down). I would have guessed that we could have offset the negative effects of age with water changes. In other words, I would have predicted an interaction between age and water changes predicting the nutrients and elements. I realize you don't have the sample size, but with a better measure for age (logged) and water changes, it might be worth a try to see if you can at least partially offset the negative effects of age with water changes.

Thanks again for doing this for the hobby.

Take care,
John

 

[rshimek]

Re: Couple of Questions

Originally posted by Experimenter [/qutoe]

Hi John,

Second, I am wondering about your assumption of normality of some of the variables. I am most interested in the tank factors, since honestly I don't really understand the chemistry or much of the biology of what is going on. In particular, I think it is probably an error to assume that (1) tank age and (2) tank size have normal distributions. Tank age, by definition, does not (it should be positively skewed). Since tank age is probably one of the most interesting factors (at least for me) I see this as a potential problem. I wonder what the correlations would look like if you performed a log transformation on tank age and tank size? It may not change anything, but you never know.

I think neither of these variables will would have a distribution that is demonstably different from normal. I may test the changes with log transformation in the future just to see, but I won't for a while.

Third, I am not sure how you measured "water changes." Is it simply "yes I do them" or "no I don't"? It would appear to me you could calculate a better measure of how much water is exchanged. Maybe such a measure would be the percentage of tank volume exchanged in an average one month period. This way you could recode the people who change 10% 4x/month to a measure of the monthly exchange (you could also do this for people who do it less often). I know this would be imperfect, but I think this issue is more important than your results show - and this could be a function of the measure and not the real factor.

In point of fact, I correlated with frequency of water changes. Additionally, this factor will be addressed more fully in next month's article.

... In other words, I would have predicted an interaction between age and water changes predicting the nutrients and elements. I realize you don't have the sample size, but with a better measure for age (logged) and water changes, it might be worth a try to see if you can at least partially offset the negative effects of age with water changes.

Again this is a factor for the next article, where I will be trying to construct a model for the build up of trace materials, but will also try to estimate changes in the tanks of all of the participants (and will send that data to them) with respect to water changes, and feeding.

 

[Experimenter]

Great! I look forward to reading more.

As for the frequency of water changes, I may have missed it in the article, but I did not remember seing good descriptions of the water change and frequency measures.

Thanks,
John

 

[rshimek]

Hi John,

I coded into the correlation analyses based on the original data. I didn't specifically discuss it in this article. More in depth in the next one....

 

[bzipkin]

I also very much enjoyed the article.
Here are a few questions for you?
Along with the details for the frequency of water changes and the percentages of actual water contents of the tank (including all resevoirs, sumps, skimmers, refugiums)
Would the use or lack of use of a skimmer (since i believe they would remove some or all of the minerals or elements talked about) change the results
I am also curious about the use (again frequency and amounts might be interesting to see)of carbon or G.A.C. and there changes in the outcome.
BZ

 

[rshimek]

Originally posted by bzipkin

Hi BZ,

[BWould the use or lack of use of a skimmer (since i believe they would remove some or all of the minerals or elements talked about) change the results[/b]

No, skimmers appear to have almost no effect. Several of the participants used skimming and it appeared to have no discernible effect.

As part of a new study, on tank exports, we will be examining skimmate and skimmer sludge, but I don't think it will have appreciable amounts of trace elements in it.

I am also curious about the use (again frequency and amounts might be interesting to see)of carbon or G.A.C. and there changes in the outcome.

Everybody in the study uses/used it. So, the levels you see are with the use of activated carbon.

 

[red hat]

chemical changes and macro algae

chemical changes and macro algae

Ron, would a larger veriety of macro algae ensure a smaller increase in trace elements and other toxins? Also wouldn't various spectrum lighting be a benefit to various colors of said macro algae. Thanks red hat

 

[rshimek]

Re: chemical changes and macro algae

Re: chemical changes and macro algae

Originally posted by red hat

Hi,

Ron, would a larger veriety of macro algae ensure a smaller increase in trace elements and other toxins?

Possibly, but I don't know. Most of these algae are competing with one another for the same resources, and even a small number of types of them may reduce the usable trace elements as far as they can. Many of the trace elements are likely to be simply poisons in the high amounts we are getting and I doubt any algal array could reduce them sufficiently.

Also wouldn't various spectrum lighting be a benefit to various colors of said macro algae.

Could be, but it depends on the algae. Where the differences in lighting spectra normally come into play are at levels of illumination far less than we normally see in reef tanks. If the tank is set to maximize coral growth, the light will be intense enough for all algae to thrive.