Mortality and survival

For any reintroduction to be successful, the new population must survive the transition to their new habitat, establish successfully and persist in the longer term. Every phase of this process has their unique challenges which however have distinct differences between departmental, scientifically followed reintroductions and community attempts at building up new colonies.

We do not have any insights into the success or failure rate of releases by rehabilitators as this seems to be a completely undocumented process. Strangely there is not even a requirement in place to disclose numbers, locations and time of release to the department. Monitoring is difficult and will not give persistence information for the actual releases if the animals are released into an established population. Then, it is just an assumption – or actually wishful thinking – that the animals seen in the area are the animals released by rehabilitators. Monitoring particularly over a longer period of time to at least show persistence of any population in a release area is usually outside the scope of those community projects. 

Official monitoring data is hard to get and most monitoring attempts of translocations have been opportunistic and without clear strategy. When funds dried up, monitoring stopped. (De Tores, 2005)

The only accessible analysis of the death rate in official research can be found in Judy Clarke’s translocation study. (Clarke, 2011) However, even in this research project there was another significant group apart from those who died or survived – those whose radio-collar stopped working prematurely.  Of 68 radio-collared animals 15 (22%) went off-air and their fate can therefore not be determined. 45 died during the 2 year research and 8 were still alive at the end of the study. These animals were not released simultaneously and being alive at the end of the study does not necessarily mean survival for a long period of time. I cannot find a definite timeframe for the study and can only assume that it started 4 January 2006 (first animals radio-collared) and ended 28 February 2008. Monitoring ceased after completion, however the radio-collars seem to have not been removed – survivors were even re-collared on the end date of the study. I assume that there were initial plans to continue monitoring and this decision was outside the influence of the researcher who was the most compassionate, empathetic and ethical person imaginable. She would not have condoned leaving ringtails with radio-collars for no good reason.

Up to the end of February 2008 5 of the survivors would have lived for 34 weeks, one for 30, one for 33 and one for 37 weeks. Their fate after this period and whether the radio-collars were ever removed is unknown.

Of those 45 that died and were retrieved only 5 deaths were unrelated to predation, however in my view this is a fairly biased statement. We do not know whether any of the other animals who were found with marks of predation were killed by the predator or whether they were scavenged. Even if they were killed by a predator, it is not unambiguous whether this animal would have or have not died anyway and was only easy prey due to an underlying cause such as malnutrition/dehydration. 

3 of those 5 deaths were attributed to hypothermia/hypoglycaemia associated with low energy reserves/bad body condition and a heavy ecto- or endo-parasite burden.
One female was found dead on the ground after a period of low minimum overnight temperatures (below 4 C) 2 weeks after release. She had lost 29% of her body weight.
Another female was also found on the ground after a cold and rainy night, 20 weeks after release.
A male was found hanging by his tail from a tree fork after heavy summer rain. He had been released only 4 weeks earlier.
A fourth mortality happened also 4 weeks after release and may have had an infectious component.  This animal had not lost any weight or body condition.
The last animal that did not display any signs of predation drowned only 2 week after release.

Criteria for success or failure according to the Translocation Proposal (De Tores, 2005) underlying this research project are split into long-term success and short-term outcomes.
In summary long-term success would have been achieved if

In hindsight we can say, the project has failed abysmally in all points but I wonder whether this was not the only possible outcome for any 2-year project. Monitoring has always been dependent on funds and there were obviously only funds for 2 years available. If the above represented the usual set of criteria for translocations, I doubt that any project could ever be deemed successful.

We have gained some very important insights though and the project has shown the interdependence of a vast number of threatening processes and the enormous gaps in knowledge that would need to be filled before any translocation could prove successful. The true value of this project can only be the increased knowledge base, but as there seems to have been no uptake of the results to improve management, this looks futile.

The (abbreviated) list of short-term outcomes is just as puzzling:

Only the first 2 objectives might have been met.  I would however argue that no 2-year project can deliver any of the remaining objectives (including those left out) and I fail to understand why new insights (not in any literature yet) would not constitute a successful result for research.

As no time frames are given that would match the 2-year research duration, I can only impose my own short-term criteria which I use to evaluate our releases.

My ‘rule of thumb’ (slightly amended for comparison with provided data):
12 weeks (3 months) – successful release
24 weeks (6 months) – successful establishment
52 weeks (1 year) and breeding – success

Following this rule, of the 45 animals that died, 20 survived the release phase (44%), 5 established successfully (11%) but only 1 would have been classified a success (2%). As it was a male, there is no chance of verifying whether he had successfully bred though.
The 8 survivors of the study would all have established themselves successfully (if my assumption of the end date is correct) which - if included in the end result - would have lifted the percentage to 52.8% and 24%. None of those animals would have met requirements for success at the end of the study. Inclusion of the 15 animals whose radio-collars failed prematurely would be for statistical reasons only as there is no evidence that their collars failed because of the animal’s death, however there is also no evidence that this was not the case. Again none had survived on-air for a year and the release/establishment percentage would only slightly change to 51.5% and 26.5%.

A few of those animals in the ‘successful establishment category’ could have already produced a recruit that was capable of surviving on its own when mother died/went off air, however the number would not be sufficiently high to sustain the population.
Alarmingly, a high proportion of translocated ringtails succumbed to predation shortly after translocation – after 2 days was the fastest.
15 lived shorter than one month after relocation (33.3 %), 10 more (22.2%) died before 12 weeks were reached. The animal that was confirmed alive for the longest period died after 61 weeks, the second longest confirmed survival lasted however only for 33 weeks.

Unfortunately there are no comparable data sets available for any other translocation of ringtail possums. I was only able to find a figure for mortality in the very first month after translocation to a predator-proof fenced site. (Hansard, 2012)
3 out of 10 died (33.3 %) in the first month which strangely is the same percentage as in the research discussed above. However without longer term figures and due to the far smaller sample size and completely different circumstances any comparison seems invalid.  All that can be deduced is that the highest mortality can be expected in the first weeks after relocation even if no introduced predators are present.

A later but now equally outdated Translocation Proposal (Williams and Barton, 2012) was drafted for the translocation of Busselton Hospital ringtails to Perup Nature Reserve but included other potential sites to be assessed for suitability.

The success criteria in this proposal are alarmingly unambitious:

No figure was provided how many females are required to have bred.

All we know about this translocation outcome is that 3 animals died in the first 3 weeks which would mean that the first criterion was met. This information was provided in reply to questions in Parliament and no further information was published since.  Anecdotal evidence points at failure of the second and third criterion. 
If not even 50% can survive for one month, the site would have to be less than suboptimal. I doubt that any translocation or carer release ever produced such an abysmal result.
That females are only expected to breed after 2 years would follow experience in inland, very low density populations. The small number of translocated females (5 radio-collared) can hardly be expected to start a viable colony. As this requirement applied to collared females only any breeding activity would have been noted as the collar/batteries would have had to be changed several times. 

Community organised releases are not comparable to any study as we cannot radio-collar the animals and therefore there is always only a very small subsample that can be identified without major doubt. Proof of survival is hardly possible without radio-collaring.
The same seems to apply to mortality as the carcasses will usually not be found, either because a predator removes their prey or because dying ringtails tend to hide under bushes or even leaf litter (personal observation). I only know of one release site used by rehabilitators where bodies were found and even a fatal fight was witnessed. My personal opinion would be that the site was seriously overstocked and competition was untenable.
In my 13 years of rehab/release work 3 animals died in the early years during the acclimatisation phase in the cage (first week) and I assume that this was caused by stress. We have amended the release protocol to eliminate/reduce the perceived problems and have not had any further deaths.

The releases described in this document (66 animals since May 2012) have not suffered any detectable early mortality, however 3 animals were found with injuries that needed treatment. This always happened in the first week after they were free. Those animals were taken back into care and all were re-released after their injuries had sufficiently healed.

One severely hypothermic pouch young was found on its own close to a release spot. The animal died after regaining normal body temperature.

Even with constant camera monitoring, we cannot estimate success or failure rates as we cannot identify most individuals, however we can prove (camera footage and night spotting) that so far at all release spots animals have persisted and bred.
If released in to unoccupied territory, individual animals can usually be followed for the period until they have bred and the offspring reached adult size. Close to 50% of our releases are identifiable in a close encounter but encounter rates are extremely low and there are only a few exceptions that can be identified on camera footage.
The longest confirmed survival for a male has so far been for 2 years, 7 months, 3 weeks and 6 days. He had been released at 7 months of age. The animal was not found dead, this was only my last close contact with this male. He had stayed in his wider release location.

The longest confirmed survival for a female has been for 3 years, 10 months, 3 weeks and 1 day. She had been released at 10 months of age. When encountered the female was with her latest offspring – still breeding at the age of 4 and a half years.

The longest survival so far would be for 4 years, 3 weeks and 4 days but only encountered when night spotting and the animal was not close enough to scan her. However, this was the only animal I ever had in care which had no white to her tail at all. The tail was actually darker than the body. As it can be assumed that this was caused by a recessive gene -  her first offspring had already a partly white tail - we are confident that this has been the original release and not an offspring. The animal was less than 2 metres from 2 observers who both confirmed that the tail was black without any colour variation. 
Interestingly this ‘unusual’ possum was also unusually strong and independent when in care and she also started eating leaves at a very young age. She was released at the age of close to 11 months. 

Apart from the ultimate cause of death – predation – some other factors were identified in the translocation study that might have increased susceptibility to predation or decreased chances of successful establishment.
Stress and disorientation seems to be the most obvious precursor of any high mortality rates in the early stages.

There was however also weak evidence that survivorship is greater for relocated animals from development sites than animals coming out of care (Clarke, 2011) of which the majority were hand-raised.
The same applied to a NSW study on common ringtails. Hand-reared ringtails survived for an average of 101 days compared to an average of 182 days by wild possums. (Russell et al, 2003, Augee et al, 1996)

Hand-reared animals are likely to have more body mass and be larger, however due to the limited cage sizes their muscle tone is often lacking and their fitness levels are lower.
The digestive health, the strength of their immune system and the training mother-raised animals receive can never be provided by a rehab facility. Hand-reared animals are neither bush-wise nor predator aware and the behavioural adaptation to captivity which can hardly be avoided might compromise the success of a later release.

Animals translocated in winter and spring also survived better than those released in summer or autumn in the translocation study.  (Clarke, 2011) Early spring when foliage quality is likely to be at its best and temperatures are mild seemed to be the most suitable time.

I personally would have liked this to be broken down into 4 seasons but even then every year has particular challenges and with climate change predictability becomes very limited.
The optimal release season is also habitat dependent. Yalgorup National Park might have provided the best nutrition in spring but one of the main ringtail predators there – the carpet python – is particularly active in spring and summer. Animals that have not settled into a new environment might be unable to withstand python predation pressure.

Also, in spring the mothers with young are highly protective, competitive and accordingly combative. I have witnessed several attacks on newcomers. Those were the main reason for released animals to be taken back into care (2 out of 3, the third cause was unknown).

Summer is clearly the hardest season for survival which again was also confirmed in the NSW study. (Russell et al, 2003)
In autumn, the food supply is suboptimal particularly after a long, hot summer or drought period and mortality rates are at their highest. Foxes are a serious threat even with a fox baiting program in place. (Grimm and De Tores, 2009)
However there is usually less aggression from other ringtails and if well supplemented during acclimatisation, ringtails can settle in before the most inclement weather starts.
In the translocation study those animals not showing signs of predation mostly died in cold, wet weather conditions. (Clarke, 2011)
If releasing into carpet python habitat, winter would provide the only definite relief from their predatory behaviour.

Ringtails that were released as adults fared better than sub-adults and juveniles. Animals above 600 grams but not sexually mature were classed as sub-adults while animals below 600 grams were classed as juveniles.
5 animals were not released – either because of their former injuries or because they were too young and small.
2 of the 3 that died in poor body condition were sub-adults when released and they died soon after.  (Clarke, 2011)
For species with complex life histories, older individuals have greater chances of successful establishment and their survival is generally higher than that of juveniles, sometimes by order of magnitude. (Canessa, 2015)
Again, the NSW ringtail study agrees that maturity is an advantage and possibly increasing survival chances.

Most rehabilitators however prefer releasing their animals at a younger age and claim that they are then more likely to be tolerated by resident ringtails.
When I tried to test this assumption those animals weighing between 650 and 750 grams were quickly moved on by their adult conspecifics instead of being tolerated. 
After weaning is complete, a sometimes significant weight loss is not unusual. The inherently stressful translocation and release process also tends to trigger weight loss. If releasing animals straight after weaning, those two risk factors in combination might jeopardise the outcome.   

dispersal issues