the online meeting place for all who love our amphibians and reptiles |
|
Evaluating population sizes and capture e |
Post Reply | Page <12345> |
Author | |
Vicar
Senior Member Joined: 02 Sep 2004 Location: United Kingdom Status: Offline Points: 1184 |
Post Options
Thanks(0)
|
Just to throw a spanner into the works, and for my own personal education. Why are absolute population figues useful ? I'd suggest..(OK playing Devil's advocate quite heavily here ), that for monitoring purposes, if the goal is to determine how well an individual population is responding, to say a management practice, why not have a sucession of similar surveys (monitoring). If sightings go up (adjusted for suitable conditions etc), then the trend is healthy, conversely, sommit's wrong. The current froglife numbers representing relative population densities could still be fit for purpose in determining key reptile sites, even though they may be arbitary figues. I say this, as I have mapped the current (old) La SAP to a proposed action plan, and actually a relatively small percentage of goals actually require population numbers, and even when required, I suspect that trend analysis would do just as well. Tho it is getting late and I'm noticing senility creeping in recently, just thought I ought to ask the 'stupid/obvious question'. |
|
calumma
Senior Member Joined: 27 Jun 2003 Location: United Kingdom Status: Offline Points: 375 |
Post Options
Thanks(0)
|
Steve, in principal what you propose is quite valid and would represent a
relative abundance that can be monitoring over time using the same methodology. However, the problem comes back down to detectability. If the management work simply alters a species detectability, any changes in relative abundance may not actually represent changes in actual population. Calculating absolute population figures are important, since they help us to determine the detectability baseline upon which estimates of population can subsequently be derived. With appropriate models, it is possible to estimate density in a way that controls for detectability. However, existing methods (e.g. capture/ recapture) require considerable effort. Chris has highlighted another important issue by reminding us that reptiles tend not to be evenly distributed across any given site. Animals tend to be patchy in their distribution. A surveyor who randomly samples a very patchy population may underestimate density (although this will be overcome with increasing effort), while an experienced surveyor who only targets focii could overestimate density for the site as a whole. Edit: I should add, that as surveyors we tend to fall in the latter category - concentrating our effort in those areas where we expect to find animals. Not unreasonable from a presence/likely absence survey perspective, but hopelessly biased when using the simple counts that are derived for estimating relative populations. Edited by calumma |
|
rhysrkid
Senior Member Joined: 14 Nov 2003 Location: United Kingdom Status: Offline Points: 98 |
Post Options
Thanks(0)
|
This is great stuff. IĈm keen to know how these ideas work with sites where animals immigrate and emigrate. With the sites that I work on that support N natrix I have been undertaking a capture-mark-recapture study (using photos of belly markings). I tend to find that some individuals stay (good reliable folk!) whilst others appear for a few weeks before never being seen again. Of course this can be down to mortality but given the species habit of having large home ranges it is just as likely that they have simply moved away from the study area. Similarly, many new individuals are seen with each survey. This could be due to high recruitment rates or alternatively immigration of individuals from other areas. When determining the population size how big does the survey area have to be in order to ascertain an accurate popn size? Would you have to survey an area that covers the maximum distance individuals would travel? If the number of individuals recorded is high is this a true reflection of the actual number relying on the site as key habitat or simply a reflection of many individuals passing through? We work within geographical boundaries such as site borders, development area or counties for example. As others have mentioned, herps care little for our way of thinking and tend to go where ever they want. Forgive me if this is not relevant, if it has already been mentioned or is straying from the main thrust of maths and models! Edited by rhysrkid |
|
Rhys
|
|
calumma
Senior Member Joined: 27 Jun 2003 Location: United Kingdom Status: Offline Points: 375 |
Post Options
Thanks(0)
|
Rhys, the most robust population estimate models should control for
emmigration and immigration. Most of the more basic techniques unfortunately assume a closed population. The problem with expanding survey areas to cover an individuals range is that you will encounter new individuals and have to increase your area to include all of their range, with the reult that you will encounter new indivi....... |
|
Vicar
Senior Member Joined: 02 Sep 2004 Location: United Kingdom Status: Offline Points: 1184 |
Post Options
Thanks(0)
|
heh ! having started to develop a model (and reviewing similar models), yes some do cater for migrant specimens. Mine certainly won't for the first few passes !...its difficult enough as it is ! What I am already finding is...even rough data, based upon extemely simple assumptions just isn't there, but then one of the 'benefits' of creating models is that it always leads to applied research requirements. Out of interest; I'm dabbling with the concept of distribution probabilites of a given population, chance of encountering, say, refugia, and detection probability. I very much doubt this will result in a method for estimating a population from survey results, but it will force me to consider,and quantify (where possible) the variables together with their relationship. This could be a complimentary approach to that being followed by Cresswell. |
|
administrator
Admin Group Joined: 01 Jan 2007 Status: Offline Points: 10 |
Post Options
Thanks(0)
|
One can't help thinking of Einstein on his death bed attempting to mathematically model the behaviour of sub-atomic particles.. Albert.. it is chaos!! I was pondering the same problem as Rhys. Consider a site that lies in the base of a valley. I survey the 40 acre area in the early spring and record a small number of adder regularly under refugia. I determine a likely hibernation area. I consult my database of records and run the numbers through my mathematical model and conclude that I am dealing with a low-density population. I plan mitigation, I erect exclusion fencing in June. During August I start to remove 8 adders a day from the site and after 2 weeks there is no sign that the numbers are reducing. My receptor site is totally inadequate I am now panicking and wondering why on earth my survey data did not reveal the numbers on site. Of course I overlooked that the site at the base of a valley was a foraging area serving 6 separate locii located on the escarpment 1km away and large number of animals moved onto the site before the exclusion fence was erected but after I completed my presence absence surveys Probably a very extreme example, but I wonder how a mathematical model could cope with this sort of fluctuation? Clearly if the surveys had extended throughout an entire season it is likely that the true nature and importance of the site to adder would have been established.. but how many consulultancy based surveys extend for the entire season? One salient point arose during my time working with mathematical models for fluid flow, rubbish in, rubbish out! Which leads to the bottom line, we need to look at standardising survey techniques, in my opinion presence absence is inadequate to base a mitigation on and not conclusive in many circumstances. Edited by GemmaJF |
|
Iowarth
Admin Group Joined: 12 Apr 2004 Status: Offline Points: 743 |
Post Options
Thanks(0)
|
Your final line sums it up Gemma - to quote Chris G-O " The absence of evidence is not the same as the evdience of absence"
|
|
Chris Davis, Site Administrator
Co-ordinator, Sand Lizard Captive Breeding Programme (RETIRED) |
|
calumma
Senior Member Joined: 27 Jun 2003 Location: United Kingdom Status: Offline Points: 375 |
Post Options
Thanks(0)
|
Gemma, for your example substitute gcn for adder and breeding pond for
focii - would that be considered acceptable? I really think that we need different techniques that adequately address the very real differences between different species. This is especially important when considering species such as adder, that may display significant temporal differences in habitat occupancy and subsequently local distribution. In the example you quote, I would argue that different population estimates should be generated (where possible), one for each component of the metapopulation. It's interesting that for amphibians, we generally only estimate populations from breeding adults in ponds. Yet for reptiles, we assume that simple counts are a true reflection of population - even when some of the same metapopulation dynamics may be at work... Alas, I also don't think that much of this debate is relevant for consultancy - at least not at the moment. The level of legal protection that widespread species receive (and consequent lack of mitigation licensing - even for adder), means that consultants can meet their client's legal requirements without worrying too much about any of this stuff. If we as practicing herpetologists (conservationists and consultants) can help to coordinate a large-scale project that attempts to address some of these issues, then perhaps we can help to influence change? |
|
Vicar
Senior Member Joined: 02 Sep 2004 Location: United Kingdom Status: Offline Points: 1184 |
Post Options
Thanks(0)
|
I suspect, in this context, it refers to gaining appropriate evidence
of the likely non-presence of a species. Not seeing a species on one
trip is probably insufficient evidence to suggest its likely absence.
This then leads to the question of how many site visits in suitable
conditions are necessary before we make the judgement that a particular
species is probably not present.
This should be a fairly easy one to crack using analysis of historical evidence. Number of visits are likely to be different by species, especially concering Ca ? But existing records should provide a distribution of number of site visits until a particular species shows up, so putting a figure on this with a degree of confidence (in general terms) should be cake compared to some of the issues on this thread :P |
|
administrator
Admin Group Joined: 01 Jan 2007 Status: Offline Points: 10 |
Post Options
Thanks(0)
|
Population dynamics and spatial distribution of the adder Vipera berus in southern Dorset, England Tony Phelps |
|
Post Reply | Page <12345> |
Tweet
|
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |