|
Post by obsidian351 on Oct 8, 2013 22:35:21 GMT 10
While cheetahs persist, haven't things like wolves died out in places like the UK? This is really not my area so maybe someone can enlighten me but my question is basically - why did they die out there? while its not always the correct info here is Wikipedias version
Wolves were once present in Great Britain. Early writing from Roman and later Saxon chronicles indicate that wolves appear to have been extraordinarily numerous on the island.[1] Unlike other British animals, wolves were unaffected by island dwarfism,[2] with certain skeletal remains indicating that they may have grown as large as Arctic wolves.[3] The species, which was a threat to livestock, human life and frequently desecrated burial sites, was exterminated from Britain through a combination of deforestation and active hunting through bounty systems.
en.wikipedia.org/wiki/Wolves_in_Great_Britain
if the above is correct it would appear it was the brit's way of dealing with things, thylacine included
|
|
|
Post by youcantry on Oct 9, 2013 10:20:11 GMT 10
Did the thylacine sweat? Or pant like a dog? I guess if you wanted to take the question one step further you'd have to ask: how *efficient* is panting/perspiration in regulating devil temperature? For example, even though panting increases 7-fold, what if panting contributes only a small percentage to thermoregulation? Then increasing sweating by 50% might have the more significant effect. There must be some maths we can propose to speculate on what percentages are required in order to make the observations feasible? For example, in order to cool more efficiently, it might be sufficient just to get a coating of sweat on our bodies, acting as a conductor for the heat. So a 50% increase might be enough to produce that coating and then the body might not continue to perspire at an increased rate. Whether or not it does continue to perspire might depend on how quickly the perspiratione evaporates - so for example, resting in the sun without wind should cause the perspiration to evaporate less quickly than resting in the sun on a windy day. Perhaps measuring rate of perspiration is actually a better measure of environmental factors affecting perspiration than it is of perspiration's ability to cool a creature. One would think there'd be studies like this on humans - understanding this stuff would have huge implications for sport, for example.
|
|
|
Post by youcantry on Oct 9, 2013 11:52:34 GMT 10
Here is my crude Thylacine population model. Hi DrTom, Would you mind spelling out how the workbook works in more detail? What do the cells represent? Surviving offspring resulting from the individual named in the column header? For example, what does "3" in cell B2 (under "M1" for year 1974) actually mean? Why is column B (M1) a sum of some cells from previous row? Why is column C (F1) just equal to B? Why are columns D & E multiplied by 0.67, but F and G are multiplied by 0.9, H and I by 0.85, J and K by 8? Why are columns L and M divided by 2? Why are N and O multiplied by 0.5? Why are the formulas only applied in the coloured cells? What does it mean for the cells above the coloured region to simply be equal to a cell from a different column? I take it the amount in the SUM column is the number of living thylacines in that year (row)? For that matter, what does "M1" and "F1" mean? I am guessing - now I've thought about all these questions, they mean "Females 1 year old", etc? I will have a play with the workbook and see if I can work these answers out, and if so, relabel things and maybe put values into visible cells (rather than visible only in the formulas).. just for fun
|
|
|
Post by youcantry on Oct 9, 2013 12:25:18 GMT 10
If I'm reading it right, you're assuming *every* breeding female (females >2yrs) produces 2 males and 2 females every productive year (ie, until their death). This seems to assume a 100% success rate (four nipples?) for breeding across every single female for nearly 40 consecutive years. I think if we reduce those figures a little we will have a far smaller population than 1,500 animals in 2013.
|
|
|
Post by youcantry on Oct 9, 2013 13:11:10 GMT 10
In fact, if I reduce the success to 50% (by changing the '2' values in the coloured cells in column E to '1'), your population climbs to 69 individuals in 1983 then crashes to 6 individuals this year.
Interestingly, there was a peak in sightings between about 1980 and 1993, at least in the north east where Tigerman predicted a population of about 40 animals.
If I add 1 individual to each gender age across your population (taking starting numbers to 36), the peak is still in 1983 and the crash is to 9 individuals this year.
It seems the peak is dependent on the spread of ages. You begin with large numbers of young individuals, then progressively fewer older ones.
If I invert the spread, the peak becomes 1981 and the crash is 8 individuals - ie they peak and crash faster.
I'm thinking you need more than 36 to keep the population alive, but I'm also thinking with your mortality selections (at the selected ages) that your model will always lead to extinction - it will just take longer for a larger starting population.
Multiplying my increased starting numbers by 10 shows a peak in 1982 of 1130 and a decline to 89 in 2013.
My conclusion is that your chosen mortality rates predetermine whether the species goes extinct or not. For a more accurate model we would need more accurate mortality rates - which I think is simply the data we just don't have, and can only speculate on. That said - it should be possible to model the rates within which the actual rate should have remained in order for any population to persist (for different starting sizes). Then we can ask (speculate) "how likely is it that these rates happened in the wild?"
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 9, 2013 17:35:52 GMT 10
Apologies for my tardiness. I threw the spreadsheet together in a few minutes the other day and just posted it, then became busy with other things. Anyway, it seems you have answered many of your own questions, which is good. I'm glad you are having fun! "Why is column B (M1) a sum of some cells from previous row?: A: All the numbers in the first line represent males (M) or females (F) of a residual population. The numbers are purely speculative. They are in an age range because in any given population not all individuals would be equal in age (unless we are down to n=1!!) Why is column C (F1) just equal to B? A: I assume equal numbers of M & F. Why are columns D & E multiplied by 0.67, but F and G are multiplied by 0.9, H and I by 0.85, J and K by 8? A: These are "survival ratios" or the opposite of "mortality ratios" and they will not be the same for all age groups across a population. For example, babies generally have higher mort rates than young adults and very old adults have higher mort rates than mid-age adults, as they become more prone to disease, infection, accidents related to frailty etc. Why are columns L and M divided by 2? A: I got lazy. A mort rate of 50% implies a survival rate of 50% and I typed in "divide by 2" rather than "multiply by 0,5". I'm lazy. And inconsistent... Why are N and O multiplied by 0.5?" A: Same reason as above. Why are the formulas only applied in the coloured cells? What does it mean for the cells above the coloured region to simply be equal to a cell from a different column? A: I tended to apply the mort coefficients when population of that cohort reached about 10. Of course, in real life it applies to all numbers, but the effect can be catastrophic in a spreadsheet when n is small. It can also be catastrophic in a population, too! Of course, I would like to include a "random death factor" to strike out 1 individual in a cell now and again to represent an accident, but I don't know how to do this in Excel. (Or even if it is possible.) I take it the amount in the SUM column is the number of living thylacines in that year (row)? Yes. If I'm reading it right, you're assuming *every* breeding female (females >2yrs) produces 2 males and 2 females every productive year (ie, until their death). This seems to assume a 100% success rate (four nipples?) for breeding across every single female for nearly 40 consecutive years. I think if we reduce those figures a little we will have a far smaller population than 1,500 animals in 2013. A: Yes, of course I have exaggerated the breeding success of the females, but I did introduce a mort coeff in Year 1 to compensate (a bit). My conclusion is that your chosen mortality rates predetermine whether the species goes extinct or not. A: Yes! That is both the fun and limitation of the models. We get to play the Deity and decide when and how the species dies! Or survives. It's always harder being an apex predator, though, I think. (Although the Passenger Pigeon may not have agreed.) We could substitute mortality rates for Tasmanian devils IF they have about the same lifespan as the Thylacine. Anyone know? PS: I'm glad you have had fun with the spreadsheet!
|
|
|
Post by youcantry on Oct 10, 2013 12:30:13 GMT 10
I've started a new spreadsheet with species that live for 3 years and tried to make it highly customisable. In particular, extending lifespan will be supported, and you can specify the proportion of females that produce from 0 to 4 young. You can specify survival/mortality rate per age grouping and age at sexual maturity. The 0-1 year is included. ie) I'm still having fun
|
|
|
Post by molloch on Oct 10, 2013 14:35:42 GMT 10
I wrote up a massive post on this yesterday but it didn't save and I don't know where it went.
To start a model off, try and eliminate as many variables as possible to give you a baseline. I built one using just these variables:
a - Age to First Litter b - Viable Offspring per litter c - Expected Litters per lifespan d - Ratio of Females to Males born e - Founding population
In Excel, set up an iterating function, and use these variables.
Population is simply Pop(n) = Pop(n-1) + Births(n) - Deaths(n)
So it is easier to set up 4 columns:
Col1 - Year (n) Col2 - Births Col3 - Deaths Col4 - Population
Variable e is really Pop(0), and to keep everything else easy, we can use a = 1 and d = 0.5, which is probably most realistic. I used c = 2, which I think is highly pessimistic, I think it would be closer to 3.
One thing you quickly notice, using the above values, is that the viable offspring value (b) has to be between 0.7 and 0.8 to keep the population alive and undetectable. (using Pop(0) = 10) less than 0.7 and the population crashes within, at 0.9 the population rises to 430,000 after 70 years. A value of 1, that is one individual per litter living to produce 2 litters, and your up to 23 million offspring at 70 years.
A couple of points: When you are creating the model use the FLOOR function in Excel to round your breeding function. We are working with actual individuals here and you can't have 0.8 of an individual breeding. You will need to base each cell of the cell above it in the spreadsheet, variables c and a will need to use the OFFSET function as they just set the start and end time for breeding. My "death" column was populated based on births at n-(a+c), you can use the "OFFSET" function in excel to do this.
These are very, very simple models obviously, but you can see the fine line an animal has to walk to stay between undetectable and extinct in a 70 year period.
|
|
|
Post by youcantry on Oct 10, 2013 15:21:46 GMT 10
What happens if you look for figures to emulate the Tassie devil? For example, estimate a starting population of about 20,000 in about 1930, peaking at 120,000 in 1990. Do we have measured data for some of these variables? If so, what value is required for (b) to produce what was actually observed (until DFTD)?
|
|
|
Post by obsidian351 on Oct 13, 2013 11:29:50 GMT 10
Well I'm at a loss to find any other population modeler made in excel, you would think someone somewhere would have done one on any other animal, I did find one on bacteria... If only Thylacines multiplied by cellular mitosis.
|
|
|
Post by youcantry on Oct 14, 2013 9:33:04 GMT 10
I'm going to take this theme (population modeling) to the new thread on that topic..
|
|
|
Post by obsidian351 on Nov 15, 2013 22:48:42 GMT 10
sitting back watching a bit of tv a few nights back I flicked over to national geographic, they was showing a doco on the Koalas of Kangaroo island, some point made where "Between 1923 and 1925, 18 koalas were released on Kangaroo Island. Their numbers increased rapidly and in 1997 a population-control program was implemented based on a population estimate of 5000 koalas" "it became clear that the koala population on Kangaroo Island was much greater and more widely distributed than previously thought, In 2000–01 the koala population size was estimated to be ~27 000 koalas" note: www.publish.csiro.au/paper/WR03007.htm it also went on to say that the low number of original Koalas made for a small gene pool and this resulted in a bunch of birth defects interesting to keep in mind when pondering the Thylacines chances of survival
|
|
|
Post by molloch on Nov 15, 2013 23:58:19 GMT 10
it also went on to say that the low number of original Koalas made for a small gene pool and this resulted in a bunch of birth defectsI was wondering where you took the photo for your avatar.
|
|
|
Post by obsidian351 on Nov 16, 2013 7:15:08 GMT 10
it also went on to say that the low number of original Koalas made for a small gene pool and this resulted in a bunch of birth defectsI was wondering where you took the photo for your avatar.
hehehe, you got me actually the birth defects were that some Koalas were born without genitalia and some were born with both or 2 sets.... That gives me an idea for a new avatar
...
|
|