I have long been an advocate of scientifically derived weight estimation as both a conservation measure and a fish recording tool.
Rack of eye estimation to facilitate the in-water release of fish is commonly used over in the US, and I have seen length and girth
measurements used for shore caught shark weight estimation in Namibia. In the UK it is now a well established tool for estimating
the weight of common skate, and was once regularly used for estimating the weights of tope. So it isn't a new concept. Having
already discussed its potential merits as a method of establishing a complimentary record fish list (see Angling Records), perhaps
its time it was looked at more seriously for other species, competitions and record keeping across the board.
With carefully prepared charts or graphs, weight estimates based on body measurements are always going to be more
accurate than weighing a fish on a scale in a pitching boat. By placing a wet rag over the face of most fish, they will
keep sufficiently still for length and girth measurements to be taken. So, with an opportunity to gather cod data through
from bait robber size to near 50 pound monsters in arctic Norway, something that would not be possible in the UK, between
drifts in flat calm conditions I noted the weights, lengths and girths of as many cod as I had time to deal with, particularly
those at the bigger end of the scale.
Using length alone to estimate weight is not an accurate option which is why I also measured girth at the widest point
of the body. Unlike Davy Holt & Bill Little's common skate estimation chart (see
www.catchalot.co.uk/tagging/tskate.htm)
which has body length along the vertical axis of the chart, and between wing tip widths along the base, I decided to multiply
length by weight to arrive at a single figure which I then plotted against body length. The fatter a fish was in relation to its
length, the greater the figure that length x girth would produce, which is another way of representing overall body size.
The advantage of this as I see it over the skate chart was that I could produce it quickly as I did not need to have
data to fill every column at every stage. A single line graph fills in the gaps for you. Obviously, having fish from every
weight category would make the graph more accurate, which is fine if you have the opportunity to get it, which I didn't. But
with a statistically predicted accuracy of 95.5% for just a few hours work, I feel it is accurate enough. In Plot 1 you will
notice one single dot out on its own at the top right hand side of the line. When recalculated with this 'outlier' removed
in Plot 2, the accuracy jumps to 97.3%. With the worst 6 outliers removed (plot 3) the accuracy jumps to 97.8%. I see no problem
in removing the single 'outlier' which was a fish of 37 pounds, but removing six could be seen as perhaps a bit too manipulative,
besides which it makes hardly any difference to the final accuracy outcome.
What I also did was remove some scales from Dave Devine's 47 pounder. Though I am trained in scale reading,
this was my first attempt at interpreting annual growth variations in cod. But the opportunity was too good to miss.
In my opinion, Dave's fish was between 8 and 9 years old which is an amazing level of growth.
Plot 1 - all data included
Plot 2 - single 'outlier' removed
Plot 3 - six 'outliers' removed
NOTE: In case anyone out there would like to continue what for me was a one shot data gathering exercise, then please feel free to add to the raw data table below.
RAW DATE TABLE
NOTE: To decimalise the weights, 1 oz = 0.065
