HOBBY GRADE TEST KITS CAN OUTPERFORM ICP MEASUREMENTS…REALLY??

[BeanAnimal]

To me the problem with all of these 'measurements' is the red, green, yellow graphs. What exactly are they based on. If a company I sent an ICP test to couldn't answer that question I would not send another

You appear to be absolutely fixated on this point. They appear to be recommending what they feel is a range close to NSW. That is a topic of its own (what target levels or acceptable ranges should be). However, the overarching issue here is that the values returned do not appear to to be all that accurate or repeatable or agreeable between vendors. The issue is not the red or green or whatever colors in the table, rather the debate is the accuracy of the underlying numbers used to create the table.

 

[MnFish1]

You appear to be absolutely fixated on this point. They appear to be recommending what they feel is a range close to NSW. That is a topic of its own (what target levels or acceptable ranges should be). However, the overarching issue here is that the values returned do not appear to to be all that accurate or repeatable or agreeable between vendors. The issue is not the red or green or whatever colors in the table, rather the debate is the accuracy of the underlying numbers used to create the table.

I disagree with your perspective and you're misunderstanding mine. The point is none of us know 'what' they are attempting to do with the ranges. The goal of sending a test to a company is to get an answer 'Can I use this value to help manage my reef tank?'. Thats the goal of hobby tests and thats the goal of ICP tests. This thread is about 'comparing hobby tests to ICP tests, right' - at least when I read the title. One comparison is how can you use the information you get From each test. If you read the title of the post again: "HOBBY GRADE TEST KITS CAN OUTPERFORM ICP MEASUREMENTS…REALLY??" I guess I would ask you 'outperform how'. In getting the absolute value of a measurement or in managing the reef tank.

 

[Randy Holmes-Farley]

I can’t see the download because my phone is full, but most labs “will list” LLOD information on their websites for ultra trace elements. Values that aren’t achievable by OES. What frustrates me is that instead of listing (< LOD) on their analysis, they will list a number (let’s say .05 ug/L. This implies to me that they’re saying they can accurately get down there which is garbage. What I’m arguing is they should do exactly what Reef Zelements is doing. For elements where OES is struggling (result is likely inaccurate or very questionable [below 2-3 ug/L for several elements]) or where OES simply lacks the “sensitivity” altogether, list that value as (< LLOD) instead of listing the inaccurate value/ noise. When people see a value there, they automatically assume they methodology was able to detect it accurately which is misleading. They need to stop that nonsense.

I agree that any method that cannot detect the analyte accurately should list the value as “not detected” or “less than X”.

 

[BeanAnimal]

I disagree with your perspective and you're misunderstanding mine. The point is none of us know 'what' they are attempting to do with the ranges.

No I am not misunderstanding you, nor do I disagree with your actual premise regarding the retuned recommendations. It just really has nothing to do with this conversation and is a broad tangent worthy of its own thread.

The goal of sending a test to a company is to get an answer 'Can I use this value to help manage my reef tank?'. Thats the goal of hobby tests and thats the goal of ICP tests. This thread is about 'comparing hobby tests to ICP tests, right' - at least when I read the title. One comparison is how can you use the information you get From each test.

No that is a different topic altogether and a pretty broad stretch of context. I think you missed the entire point of the thread. In any case, it is a free world and a free open forum and you can surely discuss whatever you want.

If you read the title of the post again: "HOBBY GRADE TEST KITS CAN OUTPERFORM ICP MEASUREMENTS…REALLY??" I guess I would ask you 'outperform how'. In getting the absolute value of a measurement or in managing the reef tank.

The context of the experiment and the title wording are very clearly regarding accuracy and repeatability of the results.As well, the overview and conclusions drawn by the authors and the bulk of the discussion are regarding the trustworthiness of those numbers with regard to their actual values vs the returned results. This is clearly not thread about discussing what trace elements are valuable, needed, dangerous or even relevant to NSW or a captive reef.

 

[BeanAnimal]

I agree that any method that cannot detect the analyte accurately should list the value as “not detected” or “less than X”.

The question for me would be (in the vendors words) why are they returning a value instead of ’not detected’ or ‘below x’. This feels like either oversight or simply reporting non-significant digits to appear more accurate and complete than the next competitor.

 

[areefer01]

No I am not misunderstanding you, nor do I disagree with your actual premise regarding the retuned recommendations. It just really has nothing to do with this conversation and is a broad tangent worthy of its own thread.

To be fair there are many posts in this thread that are not related to post #1. Even my reply now

On a more serious note hope your day, and others, is going well.

 

[Randy Holmes-Farley]

The question for me would be (in the vendors words) why are they returning a value instead of ’not detected’ or ‘below x’. This feels like either oversight or simply reporting non-significant digits to appear more accurate and complete than the next competitor.

Are you sure any vendor is giving a number below their lower limit of quantitation?

In my case, Triton did not. They said not detected.

 

[ingchr1]

Are you sure any vendor is giving a number below their lower limit of quantitation?

In my case, Triton did not. They said not detected.

Based on my thread, I would say yes. It appears that Triton and ICP-Analysis.com are reporting zeros when maybe they should be reported as non-detectable?

ICP Labs - Are the Reported Results Geared Toward the Hobby

In another thread the question of significant/insignificant decimal places reported in an ICP analysis was brought up. This raises the question that even though these labs are testing specifically for the hobby, are the results they are reporting actually geared toward the hobby? What level of...

www.reef2reef.com

 

[Reefahholic]

Based on my thread, I would say yes. It appears that Triton and ICP-Analysis.com are reporting zeros when maybe they should be reported as non-detectable?

ICP Labs - Are the Reported Results Geared Toward the Hobby

In another thread the question of significant/insignificant decimal places reported in an ICP analysis was brought up. This raises the question that even though these labs are testing specifically for the hobby, are the results they are reporting actually geared toward the hobby? What level of...

www.reef2reef.com

Not only that, but they will report a number under 1 ug/L for something like Mn, Cr, Cu, etc….all of the ultra trace elements. OES just can’t get down there for most, and anything under about 2 ug/L for the majority of those elements isn’t accurate. That’s incredibly frustrating. A few elements have better sensitivity than others, but I still wouldn’t dose based on OES results even for Co, V, Zn, etc….

 

[Randy Holmes-Farley]

Not only that, but they will report a number under 1 ug/L for something like Mn, Cr, Cu, etc….all of the ultra trace elements. OES just can’t get down there for most, and anything under about 2 ug/L for the majority of those elements isn’t accurate. That’s incredibly frustrating. A few elements have better sensitivity than others, but I still wouldn’t dose based on OES results even for Co, V, Zn, etc….

Well, that’s an exaggeration according to Triton who claims they can get below 1 mg/L for some trace elements, as I note above.

 

[BeanAnimal]

Are you sure any vendor is giving a number below their lower limit of quantitation?

In my case, Triton did not. They said not detected.

Hi Randy -

No I am not sure of that. I responded in context to Reefaholic's comments and your response, as well as the fact that many of the lower end values are so erratic between tests that the numbers don't feel to be accurate regardless of what the stated sensitivity is.

 

[Randy Holmes-Farley]

For example, when I checked in the context of my Triton test, Triton claimed a limit of detection for manganese of 0.17 ppb.

Thus, I’d be wary of generalizations of who can detect what at what level.

 

[MnFish1]

The question for me would be (in the vendors words) why are they returning a value instead of ’not detected’ or ‘below x’. This feels like either oversight or simply reporting non-significant digits to appear more accurate and complete than the next competitor.

Are we sure that the study above (the OP) actually discusses this? This post only refers to ICP companies as compared to hobbyist tests, right? In reality - the OP - does not discuss any of this. Instead we should be (I think) - be discussing whether an iodine test from ICP matches a hobby test from company xx.

 

[BeanAnimal]

Are we sure that the study above (the OP) actually discusses this? This post only refers to ICP companies as compared to hobbyist tests, right? In reality - the OP - does not discuss any of this. Instead we should be (I think) - be discussing whether an iodine test from ICP matches a hobby test from company xx.

The accuracy and precision of the values returned by the tests is the topic. So yes, I am sure.

 

[MnFish1]

The accuracy and precision of the values returned by the tests is the topic. So yes, I am sure.

The topic is the title of the post - I would think. Since only a few of the multiple tests available in ICP results are available as hobbyist tests. it's hard to state what is being said. I am curious though, where in the results do you see calculations of the accuracy and precision of the various ICP tests (i.e. the individual company's results) - I perhaps missed it.

 

[BeanAnimal]

The topic is the title of the post - I would think. Since only a few of the multiple tests available in ICP results are available as hobbyist tests. it's hard to state what is being said. I am curious though, where in the results do you see calculations of the accuracy and precision of the various ICP tests (i.e. the individual company's results) - I perhaps missed it.

You are either being insanely obtuse here for the sake of petty argument or honestly confused contextually. Either way, there is no need to sort it out in public.

As for the validity of the recommendations returned by ICP vendors, I would (honestly) suggest starting a thread, as it is a very valid topic deserving its own focus and I too question how they arrive at their suggested values and ratings.

 

[Reefahholic]

For example, when I checked in the context of my Triton test, Triton claimed a limit of detection for manganese of 0.17 ppb.

Thus, I’d be wary of generalizations of who can detect what at what level.

Kinda like these below. HeHe

I know they were in there but neither were detecting them at the levels they were at on MS.

 

[BeanAnimal]

@reefaho

Kinda like these below. HeHe

I know they were in there but neither were detecting them at the levels they were at on MS.

But again, how do we know that the MS value is correct and/or the standard to be judged against?

To make that leap you have to ignore the data presented in the opening pages of this thread.

 

[jda]

You just cannot get into the head of a fanboy, but I admire your persistence in trying... like trying to tell a swiftie that Travis Kelce being a multiple time all-pro, multiple super bowl winner, etc. did not need Taylor to be somebody. Some folks just are wanting to get it.

 

[Randy Holmes-Farley]

Kinda like these below. HeHe

I know they were in there but neither were detecting them at the levels they were at on MS.

That large table is exactly what one hopes is reported, right? It's not pretending to get values below their limit of detection. They do not suggest dosing any of those that they believe are below their detection limit that I can see.