I have finished tagging now that my research permit has expired, so I have “closed off” my data and can begin final analyses. Recaptured shells with their operculums, gonad samples and a raw data spreadsheet have been provided to NSW Fisheries for long-term curation and possible further analysis.
I will occasionally continue diving, under my recreational fishing licence. My permit allowed me to take undersize snails (less than 75mm) and to exceed the bag/possession limit of 20 snails per day. Now as a recreational diver, I can only collect legal-sized snails with tags, but I can re-tag them and possibly tag some more. This will yield more growth data around the legal minimum length (and the average maximum size), so the data shown below may be added to over time. I am unlikely to find more than 20 legal sized snails with tags at a time so I won’t be “taking” snails illegally as defined by the legislation. While the tags can be overgrown with epibionts (animals and plants attached to the shell), many shells are fairly clean and those tags are easily recognisable underwater. (see Figures 1 and 2 below)
Recognizing tags underwater on an overgrown shell (above) is unlikely, compared to the snail below. This was the reason for bringing all snails aboard or onshore to check closely for tags. It was also a reason for double-tagging and triple-tagging to maximise tag recognition. Shorter times at liberty reduce the non-recognition problem.
Below is a summary of Lunella torquata tagging data. A summary of all my length measurement data will appear in a later blog post where I will discuss length frequency analyses. Summaries of Turbo militaris data will also be in later blog posts about that species.
TABLE 1. TAGGING AND RECAPTURE DATA SUMMARY FOR Lunella torquata FROM APRIL 2018 TO APRIL 2024 AT 10 SITES.
SITE | Number tagged #1 (multiple taggings in brackets) | Single re-captures and recapture rate #2 Multiple tag re-captures (in brackets) #3 | Minimum, Mean and Maximum sizes at tagging of recaptured snails ( L1) | Minimum, mean and maximum sizes of recaptured snails (L2) | Minimum, mean and maximum months at liberty (dT) |
All data | 10,990 | 1158, 10.5% |
|
|
|
Congwong Bay 1 C1 | 662 | 70, 10.6% | 34.5,58,80 | 42.5,66.8,85 | 3.1,8.2,32.4 |
Congwong Bay 2 C2 | 992 | 115, 11.6% | 28,59,79 | 46,67,81 | 3.3,7.2,26 |
Congwong Bay 3 C3 | 2014 | 194, 9.6% | 32,55.6,88 | 37,62.3,89 | 0.5,5.1,30.9 |
Kurnell 1 | 1436 | 118, 8.2% | 23,58.6,90 | 40,64.3,90 | 0.5,6.8,35 |
Kurnell 2 | 317 | 35, 11.0% | 20,57.8,75 | 39,64.8,78 | 1.9,8.6,47.1 |
Yarra Bay | 917 | 73, 8.0% | 30,62.5,90 | 44,67.7,91 | 1.8,5.8,22.4 |
Little Bay 1 | 1453 | 120, 8.2% | 33,56.1,85 | 39,61.6,85 | 1.3,6.3,25.2 |
Little Bay 2 | 882 (132) | 164, (17) 18.6% | 27.5,49.8,77 | 35.5,57.7,79 | 1.4,4.3,14.2 |
Mahon Pool | 950 (24) | 140, (2) 14.7% | 16,49.4,83 | 25,55.3,84 | 1.2,4,28.8 |
Kioloa | 1367 | 130, 9.5% | 42,67.7,89 | 49,72.9,90 | 1.9,7.3,24.4 |
Notes: Recapture data only include “usable” recaptures and do not include recaptures where transcription errors resulted in uncertainty about size or date of tagging, where the recaptured snail was dead (an empty shell), or if the average monthly growth increment was above 5mm (suggesting some sort of data error). In all, there are 1209 entries in my recaptured snail list, but 51are not used in my analyses. Most entries (but not all) relate to a single snail.
Note #1. This includes all snails tagged including some multiple taggings at Little Bay and Mahon Pool (ie snails recaptured with a tag which were re-tagged with a different coloured tag and returned to the water).
Note #2. “single recaptures” means any usable recapture where a length at tagging and a length at recapture and both dates could be determined. They include sequential recaptures of a snail which was re-tagged and returned to the water. A single recapture data set is L1,L2,T1,T2. Recapture rate is the number of such recapture data sets divided by the number of times a snail was tagged and returned to the water, including multiple taggings.
Note #3. “multiple recaptures” are the number of usable tag-length/tag-recapture pairs where a single snail was tagged more than once then subsequently recaptured. This only occurred from May 2021 at Mahon Pool and June 2022 at Little Bay where my tag recapture rates were high. (This high rate was surprising at Little Bay which has significant recreational harvesting activity.) I did this to provide a small subset of multiple tag/multiple recapture data. Statistically, including these 19 data point gives these snails more weight in the analysis than snails which were only tagged once and recaptured once. The small degree to which any analysis might be confounded is indicated by the low number of multiple recaptures. For these multiple recaptures, only the lowest L1 and the highest L2 were used – the intermediate lengths and dates were only used for two separate, specific analyses. One was to estimate measurement error (resulting mean = 0.443mm). The recapture rate for multiple taggings was 12.2%. Only 2 snails were recaptured after being tagged 3 times.
Figure 3 below shows a snail that has been tagged with red rectangles, then upon recapture, re-tagged with orange rectangles, then tagged again on a second recapture date, with blue rectangles. The data used for such a recapture are the length and date when red tags were attached, and the length and date at final recapture (lengths for orange and blue tags were ignored).
Figure 4. Distribution of Lunella torquata recaptures for the months at liberty shown. (Frequency function in Microsoft Excel.)
There are several issues associated with “time at liberty”. Some will be covered in future blog posts about results but one interesting aspect is leaving small snails out there too long. If double or triple-tagged, it is more likely that anterior tags will be lost. It is also likely that the ventral tags (which generally last longest) will be covered by new nacre as the snail grows in its spiral pattern. Figure 5 below shows the third, ventral red tag almost covered by nacre. This can compromise the estimation of length at tagging on small snails, especially if one or two of the dorsal tags are lost.
My research was always meant to be a simple study with basic research questions:
- What is the growth rate at various sizes?
- Does it vary seasonally?
- Are sites different?
- Does growth vary between years?
- Are the two species different
- Are males and females different?
This summary of recaptures should be viewed in that context.
As to the last question, male and female recaptures are critical. Analysis of early data (Kienzle, Broadhurst and Hamer, 2022) showed very clearly from 100,000 Monte Carlo simulations that males and females grew differently (272 males, 184 females). Males had higher absolute and relative growth rates and a higher average maximum size (Linfinity). Not surprisingly, the two simple linear analyses below of the final data set (464 males, 379 females) seem to generally confirm differential growth for Lunella torquata, but not exactly the same pattern.
Figure 6 below shows Gulland/Holt plots (Gulland and Holt, 1979) for all sites combined. The regression coefficients (and therefore the growth parameter estimates) seem visibly different for females (6a) and males (6b), but again, not exactly as reported by Kienzle et al (2022). Future analysis of variance or log-likelihood estimates will determine whether the difference is significant and may clarify any growth rate/maximum size differences.
Similarly, Figure 7 below shows Ford/Walford plots for snails at liberty for approximately one year (11 to 13 months)(see blog post 4 on Growth Rate Issues) for all sites combined. They also seem to indicate a difference between males and females.
Accordingly, nearly all future analyses of these recapture data addressing the above research questions need to be done separately for males and females. A basic summary of the sex data is shown in Table 2 below. The mean increment per month for males (1.36mm) was larger than for females (1.17mm). This might be related to declining growth rate with size because males had slightly larger sizes, but could also be due to slightly longer times at liberty ( see also Kienzle,Broadhurst and Hamer, 2022 – Fig 3). Simple linear analyses can be very misleading so finer scale partitioning will be needed and will be reported in relevant future posts, following more sophisticated, non-linear analyses.
TABLE 2. RECAPTURE DATA SUMMARY FOR MALE AND FEMALE Lunella torquata FROM APRIL 2018 TO APRIL 2024 AT 10 SITES.
SITE | NUMBER OF FEMALES | NUMBER OF MALES | NUMBER WITH UNKNOWN SEX OR JUVENILES OR NOT SEXED #1 (U,J orN) | MEAN LENGTHS AT TAGGING AND RECAPTURE (L1,L2) AND MEAN MONTHS AT LIBERTY (dT)
| |
|
|
|
| FEMALES | MALES |
All data | 375 | 462 | 305 | 56.8 62.5 5.8 | 58.6 65.3 6.4 |
Congwong Bay 1 C1 | 12 | 42 | 16 | 57.4 63.9 6.5 | 60.0 67.7 7.9 |
Congwong Bay 2 C2 | 35 | 59 | 21 | 61.6 67.4 6.2 | 60.2 68.8 7.7 |
Congwong Bay 3 C3 | 73 | 91 | 29 | 53.5 60 4.8 | 57 63.8 5.9 |
Kurnell 1 | 55 | 46 | 17 | 58.7 63.7 6.6 | 57.2 64.2 6.8 |
Kurnell 2 | 16 | 12 | 5 | 60.9 66.4 9.1 | 58.3 65.8 8.0 |
Yarra Bay | 30 | 27 | 15
| 64.5 68.9 5.4 | 59.5 65.9 6.8 |
Little Bay 1 | 62 | 37 | 21 | 55.8 61.3 6.6 | 56.0 61.0 6.5 |
Little Bay 2 | 22 | 15 | 127 #2 | 47.3 57.5 4.7 | 51.1 58.7 4.1 |
Mahon Pool | 35 | 57 | 46 #2 | 44.3 49.5 3.1 | 51.6 56.9 4.3 |
Kioloa | 34 | 76 | 20 | 67.9 72.3 6.4 | 67.8 73.5 7.7 |
Note #1 : Initially, non-female and non-male recaptures were recorded as U(nknown). Most were snails whose gonads could not be extracted after boiling. Later, J(uveniles) were recorded separately, and very recently, (Note #2), recaptures which were re-tagged were recorded as N(ot sexed).
To return to my earlier comments about only working in the future on legal-sized snails, it is probably not necessary to have tagged snails at liberty for long periods of time (more than 9 months) to get useful results. Further analysis to be reported in coming blog posts will indicate whether this is a valid hypothesis. One benefit of times at liberty of less than 6 months is that empirical seasonal growth analysis using simple General Linear Models is more conclusive (watch for a future blog on seasonal growth.)
References
Gulland J A and S J Holt. 1959. Estimation of growth parameters for data at unequal time intervals. J.Cons.Int.Explor.Mer 25(1):47-49
Kienzle M, M Broadhurst and G Hamer, 2022. Bayesian estimates of turban snail (Lunella torquata) growth off south-eastern Australia. Fisheries Research : 248, 106218.) https://doi.org/10.1016/j.fishres.2021.106218