I’ve had a lot of trouble with my oceanography project. After my first two research cruises my advisor told me this was probably because I needed better technique. But during my third cruise I came up with a way to prove that no, it was not my technique; we were just operating very close to our limit of detection. (I like to tell people that my readings were something along the lines of “seven plus or minus fourteen”.)
Dr. Hand-Waver’s response to this was to send me to sea in a higher-nutrient area, where there would hopefully be more biological activity, and therefore more to measure. And there was. Now my numbers were more along the lines of 16 +/- 14. Which was an improvement–but, still, not as publishable as we’d like.
Next Dr. Hand-Waver set DannyBoy the task of trying adapt a methodology he’d been using in freshwater, in the hopes that I could use it at sea. Would it be more sensitive than my current method? Alas, the answer was no. And so we threw up our hands and shelved the project for a few months, since I was busy working on a couple of other projects.
A couple of weeks ago, DannyBoy emailed me with a draft of his paper-in-progress. He told me to look at the methodology section, and to write blurbs where necessary. (Much of the field work he’s writing about is a two-person job, and I was the second person.) The first note he left me said, “Compare to these two references; note any difference in methods.”
One of the references was our predecessor A’s paper. The other was the original original method paper. What do I mean by “original original”? Well, the reaction we use to quantify our stuff was first developed by a Dr. B–his paper is the “original original.” Most people, however, refer to a secondary paper by Dr. W, who came up with a much more convenient way to use the reaction in the field. Most people cite both papers, but it’s clear from reading the literature that they are primarily basing their methodology on Dr. W’s work.
I’d read both papers before, at this point, but I almost always referenced Dr. W rather than Dr. B. But now, here was DannyBoy telling me he wanted to cite Dr. B rather than Dr. W. So, with a groan, I went back to re-read Dr. B’s paper for the first time in three years. How did our methodology differ? Well, for one thing, we were using Dr. W’s method of detection. Trying to leave no stone unturned, I crunched the numbers, just to make sure our reagent concentrations were the same; I did the calculation before looking at Dr. B’s numbers, just to make sure the math was unbiased.
When I looked back in the numbers in Dr. B’s paper, it hit me. He said that 1 uM was a perfectly good reagent concentration–for most samples. But if you wanted to measure really really small amounts, you should bump up the concentration, perhaps as high as 10 uM. I’d never used anything but 1 uM reagent. (This is fine in freshwater, so my experiments with DannyBoy went perfectly; but everything is much more dilute at sea.)
See, I’d only really paid attention to Dr. W’s paper, in which he said that the best way to adjust sensitivity is to bump up the pH–which I’d already done at sea; I’d bumped it up as high as it could go, and it still wasn’t sensitive enough.
So I ran an experiment yesterday. Dr. W was right in a way; by changing the pH, you can boost the sensitivity by maybe 1000x, whereas by changing the reagent concentration I only managed to affect the sensitivity by 5x. But in combination with the high pH, that 5x may be exactly what we need to measure Our Stuff.
So what’s ahead? Probably a week at one of the Famous Institutes of Oceanography (we have connections with both SIO and WHOI)–somewhere with easy access to seawater, so I can test the method. (Yes, you can order seawater and have it shipped to your lab–but it’s all filtered, and I need stuff with microbes in it.) And then, if that goes well, a couple of weeks at sea to get some real measurements…probably in conjunction with SIO or WHOI again. (Dr. HW’s preference is for the Pacific, but we may end up in the Atlantic.)
The upshot is, I may finally have solved the methodology issues on this project, which was the last big hurdle I needed to get past. From here on out it may all be data gathering!! I sure hope so.