The idea behind running 32, 6-hour net sets in random locations was something that I started working on a few years ago. The traditional way to sample our reservoirs is with overnight gillnet sets, in which the nets are always set on shore, extending out into the lake, perpendicular to the shoreline. If my goal is to know what the trend in the lake trout population is, and have a good amount of statistical confidence in the data, setting a small number of shoreline gillnets (like 6 of them) overnight does not do a good job answering that question. Now, I have a set of 32 independent samples that tell me a lot more about the lake trout population. There are a lot of different statistics that I get out of this sample design, but the one that should tell us the most about trends in lake trout density is the average number of lake trout captured per net.

This is only the third year that I've sampled the lake in this way. In 2011, the average lake trout catch was 6.4. In 2012, it was 5.3. Right now, with only half the netting done, it's 7.5. We don't have a way to directly measure the actual number of lake trout in the lake, but 5 or 10 years from now, if the average catch is twice, or half, what it is now, we can safely assume that the density of lake trout in the lake has either doubled or declined by half (with a certain degree of error in these estimates, of course).

I keep using the word "density" for a reason. Even if I run this survey at the same time every year - or at the same water temperature - the biggest variable that I have no control over is the volume of the reservoir. So, for instance, this year on our first day the water was at 8233', which is 30 feet lower than it was when we did this in 2012. So even though I'm setting nets in the same locations (mostly - some of my locations are out of the water and so I have to use alternate ones), those locations are all 30' shallower than a year ago. So it changes the distribution of the fish, but it also changes the density. If you have the same number of fish and put them in a smaller volume of water, the catch rate should go up even though the number of fish is the same - because the density of fish is higher. So, when I get into the analysis of this data later on, I'm going to experiment with correcting for reservoir volume at the time of the netting, and see if that yields numbers that make sense.

So here are a few photos, courtesy of Mike Kline. Here's the biggest fish we've picked up so far, weighing in at 20.5 pounds:

This fish did have a fresh meal in its throat, so we investigated:

It was a big white sucker. Interesting. Here is what we know about Granby right now: The prey base for large macs has become increasingly sparse the past couple years. The kokanee population has taken a huge dive, and we're not stocking as many rainbows as we used to. We know that suckers are definitely not the preferred food item for lake trout, but obviously if that's the only thing available to them, that's what they'll eat.

I've been thinking a lot about body condition in these lakers. In every other fish population I deal with, when prey is scarce, you can see it very clearly in the body condition of the fish. For comparison, if you look at all the brown trout larger than 14" in a certain reach of the Colorado River, they tend to be all skinny or all fat. There's not that much variation in the body condition. What I'm seeing right now at Granby is that some of the lake trout look skinnier than any lake trout that I've handled to date. But some of them (like the one above) are in perfectly fine body condition. Here's a good example of one of the skinny ones we've seen:

That is a very skinny lake trout. Nearly starving, I would even venture to say. So I guess what I'm saying, is that under conditions of sparse forage, it appears to me that the

**variability**of body condition in large lake trout in Granby increases. It seems that a portion of them successfully switch to other prey items, and some of them don't seem to do that at all. I don't know if that seems like a relevant, much less profound, conclusion or not, but I can say that this is not the way that any other trout population that I've seen behaves. It's fascinating, and I wish I knew what determined which fish are successful and which ones waste away. When I get the data worked up, we'll see if my hypothesis about increased variability holds true.
Hi Jon, I am curious about your experimental methods and data analysis. I have a number of questions here so thanks in advance for taking the time to answer. I see that you have been trying to place your nets in the same location year by year so you can perform matched pairs analysis. Do you also place your 32 nets at the same time of day? I know you stated that your net locations were randomly selected, but you must use some guidelines for choosing a location. How do you pick the netting sites, in other words do you try for specific or varied depth ranges, look for specific or a variety of lake structures etc. If so, do you you also perform analysis to identify the depths and structures that seem to attract the most fish or even compare which net locations are the most productive? What kind of software will you be using to analyze the data? What statistical tests do you plan to perform? Along with comparing Catch Rate per Cubic Ft of Water, will you be comparing the size and body condition from year to year? Over time it would interesting to find a correlation between lake level and number, size, quality of fish caught in your nets. Do you also set nets for lake trout forage fish? That could allow additional correlations between availability of forage and the same factors.

ReplyDeleteSo haven't seen anything new lately. Guessing summer is your busiest time out of office. Any word on the low water affects on this years Kokes?

ReplyDelete