How Long to Search for a Pet Lost in a Disaster: What the Numbers Measure, and What They Do Not

Talk to Someone Now

These three lines are open 24 hours a day, every day. Every number and every phrase below was read on that organization’s own website on August 20, 2026.

Disaster Distress Helpline. SAMHSA’s page for this line, marked Last Updated: 09/26/2025, says: “Call or text 1-800-985-5990.” In Spanish, the same page says: “Llama o envía un mensaje de texto 1-800-985-5990 presiona “2.”” For Deaf and Hard of Hearing ASL callers it says: “Please text or call the Disaster Distress Helpline at 1-800-985-5990 using your preferred Relay provider.” SAMHSA describes the service as “available 24/7 to all residents in the U.S. and its territories who are experiencing emotional distress related to natural or human-caused disasters” and states: “There is no need to give any identifying information when you contact DDH.” The sentence after that one is worth reading too, so a question at the end of the call does not surprise you: the page says a counselor may ask for some basic information at the end of the call, that those questions are optional, and that they are intended to help SAMHSA keep track of the types of calls it receives.

988 Suicide and Crisis Lifeline. Call 988, text 988, or chat online at 988lifeline.org. Its site states: “The 988 Lifeline is available 24/7/365. Your conversations are free and confidential.”

Veterans Crisis Line. Its site publishes “Dial 988 then Press 1”, “Text 838255”, and a chat option, describes itself as “24/7, confidential crisis support for Veterans and their loved ones”, and states: “You don’t have to be enrolled in VA benefits or health care to connect.”

Talking to a licensed mental health professional about this is a reasonable next step, and it is one you can take at any point, including now.

A pet-loss support line is a different and narrower thing than a crisis line, and Cornell states that limit in writing about its own hotline. Tufts publishes no such statement on its helpline page, read August 24, 2026, so treat that boundary as Cornell’s rather than as something both schools have said. The ones this page routes to, with their published hours and their own limits attached, are in the section on where to talk to somebody further down. The block above is repeated at the bottom of this page.

If what you need is practical disaster help rather than counseling, SAMHSA’s Disaster Distress Helpline page also directs readers to “Call 211 for information about disaster-related evacuations, shelters, food and clothing distribution, volunteer opportunities, and other resources and referrals. Or visit the national 211 Call Center website to find the 211 information and referral center nearest you.”

What This Page Is, and What It Will Not Do

You typed a question with a number in it. How long. How many days. What are the odds now, at three weeks, at two months, at the point where the people around you have stopped asking.

There is a real answer to part of that and it is not the part you were looking for. Here is what this page does and does not do, stated first, because a page about this subject can fail in specific ways.

It will not tell you when to stop. Not as a number, not as a hint, and not as a gentle suggestion dressed up as statistics. The reason is checkable rather than sentimental, and you should hold us to it: no source we opened publishes a stopping point for a pet lost in a disaster, and this site has no credentialed veterinary reviewer and no clinical or counselling authority of any kind. A stopping rule invented on a web page would be a stopping rule with nothing behind it, handed to a stranger, about their own animal.

It will not tell you how to feel, or how you are doing. No stages, no timeline, no exercises, no instruction to accept anything. Nobody writing this has met you.

It will publish the numbers whole, including the parts that are hard to read. The recovery figures that circulate exist, they are real, and they are routinely quoted with the qualifier chopped off. A number with its qualifier removed is worse than no number, because it reads as certainty. So the figures below arrive with their study design, their sample, their limitations and their scope attached, in the researchers’ own words, and the single most important thing attached to all of them is that not one of them was measured in a disaster.

It will separate what a source states from what this site concludes. Where this page puts two sources together and the join is ours, it says so in that sentence, in the first person, so you can discount it if you disagree.

Every phone number carries the date it was read. All of them were opened on the operating organization’s own current page on August 20, 2026. A number found on a directory, an aggregator or a state government mirror does not appear here, because that is how dead helplines survive: an operator retires a line and the copies keep publishing it for years.

The Honest Answer to the Question You Asked

There is no published recovery curve for pets lost in a disaster.

That is the finding, and it is worth stating as a finding rather than sliding past it, because the absence explains almost everything confusing about this subject. Every figure of the form so many percent found within so many days that you will encounter, including every such figure on this site, was measured from ordinary lost pets: a cat through a window, a dog out of a gate, in a neighborhood that was still standing on the day the data was collected.

Here is what we did to establish that, so you can judge the claim. On August 20, 2026 we searched the peer-reviewed literature through Europe PMC for time-to-recovery data on displaced companion animals; we went back through the disaster-response best-practice literature this site already draws on, which is the material published for the agencies that run these responses; and we read the published pages of three organizations that actually ran disaster animal recoveries, in Louisiana after Hurricane Katrina, in Butte County after the Camp Fire, and on Maui after the Lahaina fire. None of them publishes a recovery-over-time curve. What they publish instead falls into three shapes, and the difference between those shapes is the subject of half this page.

That is a statement about what those searches returned. It is not a claim that no such work exists anywhere, and if a reader knows of one we would rather publish it than our own absence claim.

So this page does something different from what you asked for. Instead of a curve, it gives you four things: what the existing numbers actually measure, what the disaster record actually contains, which of the two your situation resembles, and what genuinely closes a file, which turns out not to be the animal at all.

Part One: The Cat Curve, Quoted Whole

The paper, and what kind of thing it is

The figures behind almost every lost-pet statistic in circulation come from one study: Huang, Coradini, Rand, Morton, Albrecht, Wasson and Robertson, Search Methods Used to Locate Missing Cats and Locations Where Missing Cats Are Found, published in the journal Animals in 2018 (volume 8, issue 1, article 5). We read the full text at PubMed Central on August 20, 2026, retrieved by direct request with a browser user agent, HTTP 200, 265 KB, with no retraction or expression-of-concern marker on the record.

Its authors describe the design in the first line of the abstract: “A retrospective case series was conducted where self-selected participants whose cat had gone missing provided data in an online questionnaire.” The methods section adds how the cases were gathered: the study was conducted “with cases (episodes of cats becoming lost, hereafter referred to as ‘cats’ or ‘missing cats’) recruited using a convenience sample, including snowball sampling.” It ran through SurveyMonkey between June and August 2016, and it was promoted by Missing Pet Partnership, a United States non-profit whose founder is one of the paper’s co-authors. The recruitment line is precise: “In total, 1210 cats were recruited from 10 June to 17 August 2016.”

Where those cats lived is worth knowing, because the study gets described in shorthand as an Australian study on the strength of its lead institution. The paper reports: “Of the 686 cats where country of residence was specified, most of the cats resided in the USA (59%), Australia (20%), and Canada (14%).” So it is a majority-American sample analysed by a team at the University of Queensland, and the country of residence was not specified for the remaining cats at all.

The curve

Here it is, in the paper’s own sentences, in the order the paper prints them. This is the passage the shorter versions cut:

“Of missing cats, approximately one-third (34%; 95% CI 31% to 37%) were found alive by day 7, 50% (95% CI 47% to 53%) by day 30, and 56% (95% CI 53% to 59%) by day 61. There was little increase in probability of being found alive after day 61. By one year, 61% of missing cats had been found alive (95% CI 57% to 64%) and this had increased to only 64% (95% CI 60% to 67%) by 4 years.”

The abstract states the same shape in one sentence: “Few cats were found alive after 90 days.” The conclusions state it a third time: “There is about a 33% chance that a missing cat will be found within 7 days, and about 56% by 2 months; few cats are recovered after that.”

Read those three passages next to each other, because they are the paper saying the same thing in three registers, and the flattening is not an aside. It is one of the study’s stated results.

There is a fourth sentence in the discussion that describes the same distribution from the other end, and it belongs here with its own subject clause attached, because the subject is a subgroup rather than the whole sample: “Of the cats found alive, almost all were recovered within the first 2 months, with approximately half recovered within the first 7 days.”

Two verbs, and why they should not be smoothed into one

The abstract’s summary sentence is “Of the 1210 study cats, only 61% were found within one year, with 34% recovered alive by the owner within 7 days.” Notice that it uses two different verbs for the two figures: found, and recovered alive by the owner. The results section uses a third construction throughout, found alive, and the figure caption specifies the statistical treatment: “Being found dead was treated as a competing risk.”

Those are not stylistic variations. In this paper, found, found alive, found dead and not found are four distinct outcome states, and the arithmetic depends on keeping them apart. A summary that renders all of them as “recovered” makes the numbers easier to read and quietly changes what they say.

Not found alive is not found dead, and the study is built on the difference

This is the single most misread property of the curve, so it gets its own section.

The study followed 1,044 cats for whom timing data existed. In that group, the paper reports: “Of these 1044 cats, 601 were found alive, and the median time they remained lost was 6 days (25th and 75th percentiles 2 and 21 days, respectively).” On the animals that did not come home alive, it says: “Of the 17 cats that were found dead, the median time they remained lost was 21 days (25th and 75th percentiles 6 and 91 days, respectively).” Seventeen. Elsewhere the paper puts that group at 1.7% of the total study cats and notes that its small size limited what could be analysed about them.

And the rest: “For the 426 cats not found at the time the questionnaire was completed, the median time they had been lost for was 365 days (25th and 75th percentiles 35 and 1096 days, respectively).”

Put those three numbers side by side and the shape of the misreading is obvious. If 61% were found alive by one year, the remaining 39% is not a death rate. It is overwhelmingly a not-found-yet rate, recorded at the moment an owner happened to fill in a questionnaire, which the paper handles as right-censoring rather than as an outcome. Confirmed deaths in this dataset are a small fraction of a small fraction.

One more internal detail, reported here rather than reconciled: the paper says 601 cats were found alive in its results section and 602 in the section on distances. We are quoting whichever sentence we use and not averaging them, because a source’s internal inconsistency is the source’s, and smoothing it over would be us editing a study.

What the study says about its own limits

The limitations section is unusually candid and it is short enough to carry whole rather than summarised.

On recall: “There were likely to have been recall errors in the reported study data as a substantial proportion of respondents (71%) had lost their cat more than 6 months ago prior to the date when they completed the questionnaire, with the median duration being 731 days”, which is two years. On who answered: the authors note that 90% of respondents described themselves as strongly attached to their pet and regarding it as a family member, a higher proportion than in other surveys of pet owners, and conclude: “The generalizability of the findings may thus be limited, e.g., the study population may have comprised people more likely to put in more effort to find their cat than the average pet owner.”

Sit with that second one for a moment, because it cuts in a direction people do not expect. The sample skews towards owners who searched hard. Whatever the curve describes, it is not the general population’s odds; it is closer to the odds among people doing what you are doing right now.

The one finding in this paper that gets read backwards

There is a result in Table 5 that a reader skimming the tables would take completely the wrong way, and the authors flag it themselves, so we are publishing both halves together.

The result: “Where any type of advertising were used or if the owner contacted a facility or sought professional help at any stage, the cat was less likely to be ultimately found alive compared to if that search method was not conducted (p < 0.001 for both; Table 5).”

The authors’ own explanation, in the very next passage of the discussion: “As these strategies are unlikely to reduce the probability of missing cats being found alive, these observed associations were probably due to selective implementation of these strategies more commonly when lost cats had not been found for some time, and thus already had reduced chances of being found alive.” They then name the design limit that prevents them from proving it either way: “As our study was retrospective, it was not possible to collect the chronology of search strategies.”

In plain terms: people put up flyers and start calling shelters after the easy recoveries have already happened, so those methods are attached to the harder cases. The study cannot see the order events happened in, and its authors say so. Nobody should read that line as evidence that filing reports hurts, and this page states that plainly because the raw table, read alone, says the opposite of what the paper concludes.

The scope sentence that matters more than any of the numbers

On August 20, 2026 we ran a machine search over the extracted full text of this paper for the words disaster, hurricane, wildfire, flood, evacuate, evacuation, earthquake and tornado. The count was zero for every one of them. The word displaced appears exactly once, in an ordinary sense, in a passage about indoor and outdoor cats.

That is the whole scope problem in one measurement. This is a study of cats getting out of houses in neighborhoods that still had houses in them. It is a good study, its authors were careful, and it does not contain your situation.

Our page on running a feeding station and a trail camera for a displaced cat works through the distance half of the same paper and makes the same point about a burn scar, and our page on whether an indoor-only cat should be microchipped carries the outdoor-access distance comparison. Neither is restated here.

Part Two: The Dog Numbers, Which Nobody Quotes

Every recovery-timing figure in the previous section is a cat figure. That is not an accident of what we chose to publish; it is the shape of the literature. The peer-reviewed recovery-timing work on lost cats is a study of 1,210 animals. We located no equivalent for dogs at that scale, and the Missing Animal Response Network’s own page on the cat study, read on August 20, 2026, still describes a dog version as a future goal rather than a completed one.

What does exist is a pair of companion papers from 2007, and they are worth putting side by side because the difference between the species is larger than most owners assume.

Lord, Wittum, Ferketich, Funk and Rajala-Schultz published two cross-sectional telephone surveys in the Journal of the American Veterinary Medical Association in January 2007. Both drew on owners in Montgomery County, Ohio, whose animals were lost between June 1 and September 30, 2005. We read both abstracts through the Europe PMC interface on August 20, 2026. The JAVMA full texts were not opened and have no PubMed Central deposit, so nothing beyond the abstract is published here.

Dogs (JAVMA 2007;230(2):211-216, PMID 17223753), sample population owners of 187 dogs: “132 of the 187 (71%) dogs were recovered; median time to recovery was 2 days (range, 0.5 to 21 days).” The same abstract records how they came back: “Dogs were recovered primarily through a call or visit to an animal agency (46 [34.8%]), a dog license tag (24 [18.2%]), and posting of neighborhood signs (20 [15.2%]).”

Cats (JAVMA 2007;230(2):217-220, PMID 17223754), sample population owners of 138 cats: “73 of the 138 (53%) cats were recovered; median time to recovery was 5 days (range, 0.5 to 81 days).” And how: “Most cats (48 [66%]) that were recovered returned home on their own or were found in the neighborhood (5 [7%]); most other cats were recovered through posting of neighborhood signs (8 [11%]) or calling or visiting an animal agency (5 [7%]).” The sentence directly after that one in the same abstract is worth carrying with it: “The highest success rate for any of the search methods that were used was only 12% (posting neighborhood signs).”

Three things follow, and the third is the one to carry.

First, the medians differ by more than a factor of two, and the recovery percentages differ by eighteen points. A single sentence telling you that pets are usually found within some number of days would be wrong about one of these species no matter which number you picked.

Second, the two species come home by completely different routes. Dogs arrive through an agency, a tag or a sign, which is to say through other people and through paperwork. Cats mostly walk back in or turn up next door. That difference is doing a lot of work later on this page, because a disaster damages those two channels unequally.

Third, and this is the one that gets misquoted: the dog range tops out at 21 days and the cat range at 81 days, and neither of those ceilings is a finding about the animal. The study window ran from June 1 to September 30, 2005, four months, and owners were surveyed by telephone about that window. A dataset cannot record a recovery longer than the time it was watching for one. Treat those upper bounds as a property of the study, not as an outer limit on a dog.

And the same scope caveat governs both papers as governs the cat curve. These are ordinary lost animals in one Ohio county in one summer. No disaster is in them.

Part Three: What the Disaster Record Actually Contains

If no curve exists, what is there? Three kinds of thing, and they are not interchangeable. A rate is a share of a group, measured once. A snapshot is a count of open files on a given day. A case is one named animal on one dated day. Putting two of them on the same axis would produce a trend nobody published, so this section keeps them apart and labels each one.

Hurricane Katrina: a rate, and a record-keeping failure

The Louisiana SPCA was the responding animal welfare organization in New Orleans in 2005 and it publishes its own account of that response. We read it on August 20, 2026.

On scale, the page states that its Japonica Street shelter was destroyed in the flooding, that a shelter was created at the Lamar-Dixon Expo Center in Gonzales, and that within days more than 7,000 addresses were on the rescue list. It reports that over 15,500 animals were ultimately rescued, and that only 15% to 20% of them were ever reunited with their owners. We are giving those two figures outside quotation marks deliberately: the page pads both of them with non-breaking spaces as a layout device, and reproducing that inside quotation marks with ordinary spaces would be altering characters inside a quotation, which this site does not do.

What that figure is: a share of the animals its response physically handled, who went home. It is not a curve, it has no time axis, and it is not a probability for your animal.

The sentence immediately after it on that page belongs here, because leaving it out would make the figure look worse than the organization presented it. The page adds that although the percentage appears low, it fares better than what it calls the national average of 10%, and then says something we would have written ourselves: that for the owners searching for their pets, percentages hardly matter. We are not republishing that 10% as a statistic, because the page cites nothing for it and we could not find what it rests on. It is the operator’s own comparison, and it is reported here as that.

What it sits next to on that page is the part that has more to teach a searching owner, and it is quotable cleanly:

“A lost pet database was created to help reunite rescued pets with owners. Unfortunately, clear documentation identifying where animals were found and transported was sorely lacking, a casualty of the City’s infrastructure being destroyed and communications compromised, the chaos of the Lamar-Dixon Expo Center and some rescue groups operating outside the system.”

And on where the animals went: the page states that thousands of animals were transported to other states by more than 40 rescue groups, because Lamar-Dixon filled over capacity and no single facility in the state was large enough. Those figures are padded with non-breaking spaces in the source the same way, so they are stated here rather than quoted.

Then the sentence with the longest reach on the whole page:

“In the coming years, several lawsuits arose when rescued pets were adopted before their owners were able to locate them. The last filed lawsuit was closed in 2014, nine years after Hurricane Katrina.”

Nine years. The same page describes the Lamar-Dixon Expo Center as having become the largest animal shelter and animal rescue operation in U.S. history, which is that organization’s own characterisation of it rather than an independent finding. What matters here is the date: the organization that ran that response records, on its own site, that files from that disaster were still legally live most of a decade later. It is not a statement about how long animals survive. It is a statement about how long the paperwork runs, which is the subject of Part Five.

Colorado’s Marshall Fire: a snapshot of open files

One peer-reviewed paper reports how many lost-pet reports a receiving shelter still had open one month after a fire. That figure, and the conclusion this site draws from it, are published on our page about filing a disaster lost-pet report, which owns the reporting and maintenance side of this subject. We did not re-open that paper for this page, so rather than repeat figures from our own live page as though they were a source, we are pointing you at the page that read it. What the figure is, categorically, is a snapshot: a count of files in one state on one day at one shelter. It says nothing about what happened to those files afterwards, in either direction.

The 2018 Camp Fire: dated individual recoveries, and one practitioner’s account

Two documents record cats recovered from Paradise, California after the Camp Fire, and they are different kinds of document.

Alley Cat Allies co-ran a recovery center for those animals with FieldHaven Feline Center and published an account of it, dated February 13, 2019 and modified February 7, 2025 by its own page metadata, read August 20, 2026. It is an advocacy organization’s storytelling page rather than research, and the cases in it are individual animals rather than a sample. As individual dated cases, though, they are exactly the thing this page needs, because they sit inside the window where the ordinary-loss curve has already flattened.

On a cat named Hunter: “Large, long-haired, and grey as the ash around him, Hunter hunkered down for 62 days near the wrecked remains of the Paradise home where he once lived.” The page goes on to describe how he was found, and the method will be familiar to anyone who has read our feeding-station guide: a volunteer set out a night-vision camera and a food station nearby, Hunter came out to eat and was caught on camera still wearing the collar and bell his family had given him, and it took another three weeks of trapping attempts before he was caught and reunited.

On a cat called Hulk, rescued on December 13: the page states he “survived for a month in what Thomas R. described as the desolate remains of a Paradise neighborhood until he finally wandered into a humane trap set next to a burned-out car”, at three in the morning, with the volunteer watching the trap from a distance.

The organization’s president and founder, Becky Robinson, is quoted on the page saying “The area was left in ruins by the wildfire, but cats are survivors” and then: “Some cats are just now coming out of hiding, and many of them have ears and paws burned by the flames.” A note on that quotation: the page’s HTML carries a stray control character immediately after each closing quotation mark, an encoding artefact rather than anything the organization wrote, so we have ended the quoted spans before it rather than reproduce or silently fix a character inside a quotation.

There is a third case on that page which is not about duration at all and is worth knowing anyway. A cat called Ellie was brought in wearing a collar tag with a name and a phone number, and the phone number was a landline at a house that had burned down. The tag was accurate and useless at the same time. What eventually reunited her was people: the recovery center posted her on the county’s rescued-animals website, asked everyone who came in from that area about the owner by name, and a local shelter passed word back that the owner had come in looking. Our guide to auditing what a microchip registration actually says is about closing exactly that gap before it opens.

Rachel Zarnt wrote a first-person account of volunteering in Paradise, published in the IAABC Foundation Journal, Issue 17, August 2020, with page metadata recording publication on July 27, 2020 and modification on March 18, 2021. Read on August 20, 2026. Be clear about what this is: a personal essay by a former veterinary technician in a professional association’s journal, carrying the journal’s own note that it won the 2020 Rebecca Park Animal Behavior Consulting: Principles & Practice Scholarship. It is not peer-reviewed research, it produces no rate, and it should not be converted into one.

What it does contain is a documented event that no study could give us:

“Two weeks after the fire, Animal Control’s official word to the community was that “no more animals have survived in Paradise.””

She records what followed: that she returned to Paradise nearly every weekend for two months, that she personally witnessed dozens of animals brought into clinics several weeks after the fire, and that she was “increasingly worried about the official stance that there were “no surviving animals,” and was surprised when rescue operations were halted only weeks after the fire”. Then two sentences that belong together and are usually separated: “Unofficial rescue operations continue in Paradise and neighboring Magalia to this day. The most recent fire cat rescue and reunification happened over 300 days after the Camp Fire.” That last sentence was written in mid-2020 and describes the position then, so read it as the state of play at the time of writing rather than a running total.

One more line from that account, and we are carrying it because it points against the thing most people assume about disasters and distance:

“However, often cats could be found within several hundred feet of their home, although no buildings were left standing for several blocks in all directions.”

Hold onto that. It comes back in Part Four.

The 2023 Lahaina fire: a first report that was wrong, and a count with no clock on it

Hawaiʻi Public Radio published an interview on August 5, 2026 with Nikki Russel, Chief of Operations at Maui Humane Society, marking three years since the fire. It is a news interview with a named officer of the operating organization, which is a different evidentiary tier from either a study or the organization’s own page, and it should be attributed that way.

Russel describes the first days:

“We were begging to get in,” Russel said. “And again, the first report is that ‘You don’t need to get into the burn zone. There’s nothing alive.’”

The article then reports what happened after officers got in about a week later: “That first night inside, they rescued around 50 cats, focusing mainly on those that were burned or otherwise injured. That number would later surpass 800. And retrieving 800 cats from a 2,170-acre burn zone was no small feat.” It records that the organization worked with ASAR Training and Response and Neighborhood Cats, that many of the cats had been free-roaming and had survived by sheltering in the sewer system, that each cat removed from Lahaina got a complete physical exam, and that “All the cats who survived were reunited with their families, rehomed, or released into areas where the community could care for them.”

A sourcing note this page owes you. We opened Maui Humane Society’s own fire-response page on August 20, 2026. It returned HTTP 200 and it publishes no numbers at all: it is a photograph gallery. The figure of 800 cats and the three-year timeline exist, as far as we could establish, only as a named officer’s statement to a news outlet. We looked for an operator-page confirmation and did not find one. Treat the number accordingly: it is on the record, from someone in a position to know, and it is not published by the organization she works for.

Two statements, two disasters, and what they are and are not

Notice what happened twice, five years and an ocean apart. In Butte County, an agency told a community at two weeks that no more animals had survived. On Maui, the first report into the response organization was that there was nothing alive in the burn zone. In both places recoveries continued afterwards.

This is this page’s own reading of those two documents, not a claim either source makes, and it is worth stating carefully because it is easy to overstate. Two documented instances are not a pattern, and neither account tells you anything about your animal. What they do establish is narrower and still useful: an early official statement that nothing survived is a statement made under enormous pressure, with almost no information, at the point when least is known, and it has twice been followed by recoveries. It is not a verdict, and it was not the last word in either place.

What each disaster number counts

Every cell in this table describes what the figure in it measures, and the sources were read on August 20, 2026 except where the table says otherwise. There is no column for probability, because none of these sources publishes one.

Figure	Source, and what kind of document	What it counts	What it does not tell you
Over 15,500 animals rescued, 15% to 20% ever reunited	Louisiana SPCA, the responding organization’s own Katrina page	A share of animals its response handled who went back to an owner, measured once	Any time axis at all. Nothing about days, weeks or months
Last filed Katrina lawsuit closed in 2014	Same page	The end date of litigation over animals adopted before owners located them	Anything about animal survival. This is a paperwork clock
Lost reports still open one month after the Marshall Fire	Peer-reviewed paper, quoted on our own lost-pet report filing page rather than re-opened here	A count of files in one state, at one shelter, on one day	What became of those files afterwards, in either direction
Hunter, 62 days; Hulk, about a month	Alley Cat Allies, an advocacy organization’s account of a recovery center it co-ran	Two named animals, on dated days after one fire	A rate. Two cases are two cases
Reunification over 300 days after the Camp Fire	Rachel Zarnt, first-person volunteer account, IAABC Foundation Journal, August 2020	The latest recovery known to one volunteer as of mid-2020	A running total, or anything about frequency
Around 50 cats the first night, later surpassing 800	Maui Humane Society Chief of Operations Nikki Russel, quoted by Hawaiʻi Public Radio, August 5, 2026	A cumulative count of animals retrieved by one organization from one burn zone	A share of the animals that were there, or the period the 800 were retrieved over, which the interview does not state. And MHS’s own fire-response page publishes no numbers

Weld any two rows together and you have invented a trend. That is a real temptation here, because a reader wants the rows to add up to an answer, and they do not.

Part Four: The Distance Question, and Which Way It Actually Points

There is a widespread idea that a disaster makes animals travel further, so a normal search radius is useless. It is half right, and the half that is wrong matters, because the wrong half sends people driving around counties while the animal is under a slab two hundred feet away.

Nothing we opened supports the claim that displaced animals travel further. The only direct observation we located points the other way. Zarnt, working the Paradise burn scar in the weeks after the Camp Fire, writes that “often cats could be found within several hundred feet of their home, although no buildings were left standing for several blocks in all directions.” That is one volunteer’s observation rather than a measurement, and what it observes is the distance assumption holding through a total loss of the built environment rather than breaking.

Two things do genuinely break, and they are both real.

The landscape that produced the distance figures is gone. The published radii were measured in places where the fences, sheds, hedges, decks and crawlspaces that shape where a cat hides were all still standing. Our page on a feeding station and trail camera for a displaced cat works this through in detail and declines to restate those radii as a disaster search radius, which is the right call and we are not going to undo it here. What survives is not a number but a behaviour: animals hide close, in the tightest concealed space available, and after a fire or a flood the inventory of tight concealed spaces has changed rather than disappeared. Zarnt’s own list of where survivors were found is a list of things that were still there: burned-out camper van shells, melted car chassis, drainage ditches, and fireplaces standing alone in the rubble.

The response system moves animals, and that has nothing to do with the animal’s legs. This is the mechanism that actually defeats a local search, and it is administrative. The Louisiana SPCA’s page records that thousands of animals were transported to other states by more than 40 rescue groups after Katrina because the intake facility filled. Our page on what happens when a lost pet is transported out of the county or the state covers how transport manifests work and which national records can still match you once the animal is nowhere near where it got loose. That is the real answer to a search area that keeps coming back empty, and it is a reason to widen the reporting rather than the walking.

So the practical version, and this join is ours rather than any source’s: search physically as if the animal is close, because the only disaster-zone observation we found says it is, and report administratively as if the animal could be three states away, because the documented record says the response system moves animals across state lines. Those two are not in tension. They are two different instruments pointed at two different failure modes.

Part Five: What Actually Closes a File

Here is the shift this page is really about. Almost everybody searching for an animal after a disaster is watching the wrong clock. They are watching the animal, and asking how long an animal can last out there. The clocks that actually end searches are on paper, they run on schedules nobody announces, and every one of them can be reset.

The operator’s own answer to the question you asked

Petco Love Lost runs one of the national lost-and-found databases, and it is the one the rest of this site routes to most often. Its help center answers this exact question. The article is titled “My pet has been missing for over a month. Is there still a chance I could find them?”, created August 19, 2025 by the help center’s own record.

A retrieval note first, because it matters for anyone checking us. Both petcolove.org and the help center’s web front end returned HTTP 403 to our automated requests on August 20, 2026. That is one route failing, not a dead source. The article bodies were retrieved successfully, HTTP 200, through the help center’s own public content API on the same host, which also returns each article’s creation and update timestamps. That is the route we used and it is where these words came from.

The answer in full:

“Yes, there is always a chance of being reunited with your pet—even after weeks or months. Families have been reunited long after their pets first went missing. That said, it’s important to understand that the more time that passes, the greater the risk that your pet may have been rehomed or adopted. Most shelters are required to hold stray pets for only a short period of time (the “stray hold”). Once that hold ends, the shelter or rescue can legally place the pet up for adoption. This is why acting quickly when a pet goes missing is so important. The sooner you create a lost pet listing and begin contacting shelters, the greater your chances of being reunited. If your pet has already been adopted into a new home, unfortunately, there may be very limited options for recovery.”

Read what the operator located the risk in. Not the animal’s survival. The stray hold, and what a shelter is legally permitted to do after it expires. That is a rule with a date on it, set by state law and local ordinance, and it varies enough that a single national number would be wrong somewhere. Our guide to proving a pet is yours and working the reclaim process goes jurisdiction by jurisdiction on hold periods, on what happens when a hold expires, and on the point at which the question stops being a shelter-counter question and becomes a legal one for an attorney licensed in your state. If the animal has already been placed with somebody else, what to do when someone has adopted your pet is the page for that, and it carries the free and low-cost civil legal help routes, with each operator’s own stated limit, for a reader who cannot pay an attorney.

That page also names a failure this page wants to head off, because it is the administrative twin of giving up: an owner is told the hold is three days, concludes on the fourth day that the animal is gone, and stops calling. Call anyway, and ask for the calendar date rather than the number of days.

Your own paperwork expires quietly

The second clock is your report, and it is the one people are most surprised by. Our page on filing a disaster lost-pet report owns this subject in full, including the hold-extension range the disaster-response best practices describe, the database deactivation rule and the reactivation route, so it is not restated here. Go and read it, and then set the reminder it tells you to set.

The one thing worth adding here, because it belongs to the long search rather than the first week: the deactivation is not silent by design. Petco Love Lost’s help center states that it sends a notification when a report has been inactive, and describes what an expired listing looks like in your own account, greyed out rather than gone, with a button to bring it back. So the failure mode is not usually the operator hiding something from you. It is an email arriving in month two, to an address you set up in a hotel, and being missed.

The third clock: the response itself demobilises

Temporary shelters close. The disaster-response best-practice literature, quoted at length on our report filing page, instructs planners to decide in advance what happens to animals that were never reunited when a shelter winds down. That is a scheduled event with a date, and it is the moment to be physically in the building rather than on the phone.

What this means for a long search

Put the three together and you get the practical shape of a search that has run past the point of daily effort. What you are maintaining is not hope, or belief, or stamina. It is four pieces of paper and a phone number:

• A lost report on file at every facility that might receive the animal, including facilities outside your county.
• A database listing that has not silently expired, carrying a photograph that still looks like the animal.
• A microchip registration pointing at a phone that rings, which is the mechanism by which a shelter three counties away reaches you without either of you having done anything clever. Our page on what a microchip registration actually says covers the case where the registry shows a breeder or a rescue rather than you.
• A dated record of what you did and who you spoke to, which is what our flyer and shelter call log page is for, and which, on our own reasoning rather than any source’s, is the record you would want if the file ever becomes a dispute.

None of those four requires you to be searching. That is the point of them.

When the maintenance itself is not available to you

A page that funnels everything into keeping the listing active owes an answer to the reader who cannot. Some of these are common after a disaster and none of them is a dead end.

No internet, no smartphone, or no account. The database listing is one channel and it is not the load-bearing one. The load-bearing one is a lost report physically on file at the facilities that receive found animals, which is made in person or by phone. Ask each facility whether it takes a written report, whether it will read your report back to you, and which database it posts its own found animals to, then ask a relative or a neighbour to file in that database on your behalf using their account and your phone number.

The phone number on everything is a phone you lost. Change it once, everywhere, in one sitting: every facility report, the database listing, the microchip registry, and the flyers. A stale number on a found animal’s tag is the Ellie case above, and it is recoverable only by other people being persistent.

Your county’s shelter does not take the species you lost. Some do not impound stray cats at all. Our reclaim guide documents a county that says so on its own page and names the alternative in the same breath. Ask each agency what it takes in, not only how long it holds it.

You have moved away from the disaster area. The reports stay where the animal is, not where you are. Keep them filed there, keep an out-of-area contact number on every one of them, and ask each facility how it notifies an owner who is no longer local.

You cannot face making the calls. Hand the call list to somebody else. A neighbour, a relative, a friend who wants to help and does not know how. The log page exists partly so the job is handable: it is a list of facilities, dates, names and what was said, and a person who has never met your animal can work it.

Part Six: What Named Authorities Publish About Duration, Quoted Separately

Four organizations publish something about how long to keep going. All four were read on their own current pages on August 20, 2026. They are printed here separately and deliberately, because merging them into one tidy framework would be composing guidance that none of them wrote, and this site has no standing to compose guidance on this subject.

Humane World for Animals, on its page about finding a lost dog, under a heading reading “Don’t give up”:

“Animals who have been lost for months (and even years) have been reunited with their owners. Persistence and a systematic approach are key.”

That is the whole of it. Two sentences, no timeframe, no rate, no schedule.

The ASPCA, on its lost-pet page, under a heading reading “Don’t Give Up!”:

“This one is important! Remember that many lost animals have found their way back home.”

Also the whole of it. The ASPCA’s page is long and detailed about flyers, neighborhoods and social networks, and on duration it says exactly that and nothing more. Do not let anyone attribute a schedule or a stopping rule to the ASPCA on the strength of this page; that sentence is the whole of what it says on the question.

Petco Love Lost answers the question directly, in the passage quoted in full in Part Five above, and it is the only one of the four that names a mechanism: the stray hold, and the risk that an animal is rehomed or adopted.

The Missing Animal Response Network is the fourth voice, and it needs a warning attached rather than a quotation. Its founder, Kat Albrecht, is a named co-author of the Huang study, and her own page about that study describes in the first person how she pressed for years for a study of this kind, was then approached by an emeritus professor at the University of Queensland who proposed they collaborate, and worked on it. So MARN cannot be cited as independent corroboration of the cat curve: that would be one source counted twice. Two other things on that page are worth recording. The organization was previously called Missing Pet Partnership and both names are still in circulation, which matters if you are checking an older citation. And the in-house statistics it describes from 2013, about cats found within a five-house radius, are self-described counts compiled by that organization rather than published research, so they should be read as that or not at all.

Notice what none of the four does. Not one of them publishes a day count, a probability schedule, or a point at which to stop. If you find a page that does, look for whose data it is standing on, because as far as we can establish there is none.

Part Seven: A Search That Changes Shape, Triggered by Events Rather Than Dates

Our post-disaster reunification workflow covers the first forty-eight hours and the point where daily searching becomes something you cannot sustain, and that page owns the taper itself. This section is about the thing underneath it: what should trigger a change in what you are doing.

The default trigger people use is elapsed time, and it is the worst available one. It has no relationship to anything happening to your animal, it has no relationship to what the sources measure, and no source we opened publishes a schedule it could be checked against.

Two better triggers exist, and both are things you can check.

Your own capacity is a legitimate trigger, and it is the only one about you that belongs in this decision. A search plan you cannot sustain is not a more determined search plan; it is a plan that will stop entirely on a day you did not choose, usually all at once. Picking a plan you can keep is what keeps the reports alive long enough for somebody else to reach you, which is a fact about files rather than a judgment about you.

Events are the other trigger, and they are the useful ones, because each of these genuinely changes what is true about your search. None of them is a date on a calendar and none of them is a countdown. They are things that happen, and each one has an action attached:

• A temporary shelter announces it is closing. Go in person before it does. Ask what happens to animals still in care and who takes custody of them, and get the name of the receiving organization.
• A facility tells you it transported animals out. Ask where, get the receiving facility’s name, and file a report there. This is the single most common reason a well-run local search finds nothing.
• Your database listing changes state. An expiry notification, or a greyed-out listing in your account, means the report has stopped matching. Reactivate it.
• Your phone number, address or email changes. Redo every report and the microchip registration in one sitting.
• The animal’s coat changes with the season, or the photograph you filed is no longer what the animal looks like. Refresh the photograph everywhere it appears.
• A new database, a new community page, or a new rescue group appears in your area. File there.
• You get a sighting, credible or not. That resets the physical search to a specific place, and our guides to a feeding station and trail camera and to an animal that will not come to you cover what to do with it.
• Access to an area changes. A closure lifts, a checkpoint moves, a demolition is scheduled. Our page on getting escorted access back in to reach an animal covers the ask, and our page on requesting a welfare check or a rescue covers asking somebody else to go instead of you.

That list is this site’s own construction, assembled from the documented failure modes on the pages linked from it rather than taken from any single published protocol. We are saying so because a checklist that looks official carries more weight than it has earned, and this one is ours.

Where to Talk to Somebody About a Search That Has No End Date

The three 24-hour lines at the top of this page are the ones to use if you need someone now. This section is about something narrower, and the distinction is one Cornell states in writing about its own hotline.

A pet-loss support line is not a mental-health crisis line. Cornell says it about its own service, and we are repeating the institution’s own words rather than paraphrasing them: “The Cornell Pet Loss Support Hotline is not a mental health hotline. If the concerns voiced over a call are beyond topics related to pet loss and pet grieving, volunteers will help you find appropriate resources for your needs.” Tufts publishes no equivalent statement on its own helpline page, read August 24, 2026, so this limit is Cornell’s, about Cornell’s line, and not something both schools have said.

Two veterinary schools run these lines. Both numbers, and both sets of hours, were read on each school’s own page on August 20, 2026.

The Cummings School of Veterinary Medicine at Tufts University runs the Pet Loss Support Helpline at 508-839-7966. Its page states: “The helpline is staffed from 6 pm to 9 pm Monday through Thursday EST and has 24-hour voicemail. Calls left on the voicemail will be returned at the next scheduled shift.” Those hours belong with the number wherever you write it down, because this is not a 24-hour line, and reaching a voicemail at two in the morning is a different thing if you were expecting one. This is the most on-point resource for the situation this page is about: the helpline’s own published scope explicitly covers separation from an animal without a death, and our page on what to do when a pet dies in a disaster quotes that scope paragraph in full rather than us reproducing it twice in different words.

Cornell University College of Veterinary Medicine runs the Cornell Pet Loss Support Hotline at 607-218-7457. Its hours are deliberately unfixed, and the page says so: “The Pet Loss Support Hotline’s hours vary seasonally. Please call the hotline for current hours.” The call is answered through Google Voice, which prompts for a name; the page says you can say anonymous or give only a first name.

The two crisis numbers on Cornell’s page, and why both of them work. On August 20, 2026 that page pointed a reader looking for resources to address thoughts about suicide, self-harming or harming others to the ten-digit number 800-273-8255, and elsewhere on the same page its resource list carried the three-digit route, listing “988 Suicide and Crisis Lifeline: Call or text 988, online chat at https://988lifeline.org/chat/”. Both of those reach the same service, and this page says so plainly, because a reader who has the ten-digit number written down somewhere should not be left thinking it is dead. The operator answers the question directly. SAMHSA’s 988 FAQ, read on August 24, 2026, asks: “Does the 1-800-273-8255 number still work?” and answers: “Yes. Although the 10-digit number was transitioned to the easier-to-remember “988” in July 2022, it remains functional and will connect you to 988 Lifeline services.” The same page states: “The National Suicide Prevention Lifeline, reached by dialing 1-800-273-8255, was relaunched in July 2022 as the 988 Suicide & Crisis Lifeline, which is reached by dialing the easy-to-remember phone number, 988.” So what changed in July 2022 was the branding and the arrival of 988 itself, not the ten-digit line, and Cornell printing the older number alongside 988 is not an error on Cornell’s part. This page publishes 988, because it is shorter and easier to hold onto at the worst moment, and it is the route we would ask you to use. If the ten-digit number is the one you already have, it still works.

Cornell’s page also carries the instruction we are repeating because it is the correct one and it is theirs: if you are experiencing a mental health crisis or feel like harming yourself or someone else, call 911 or your local emergency service.

Your Long-Search Checklist

This is the maintenance version, for a search that has already run past the first week. The first-week version, in order, is on our post-disaster reunification page, and the report-filing mechanics are on our lost-pet report page. Nothing on this list is keyed to a number of days, because none of the sources supports one.

• Confirm a lost report is physically on file at every facility that could receive the animal, including facilities outside your county, and get each one to confirm it is still active rather than assuming it is.
• Ask every facility the transport question, in these words: where did your overflow go, and were any animals transported out. Write down the answer and the date.
• Check your database listing is not greyed out or expired, and reactivate it if it is. Watch for the expiry notification email, and check the folder it will land in.
• Confirm the microchip registry has a phone number that rings and a second contact who is not in the disaster area.
• Refresh the photograph on every listing if the animal’s appearance has changed with the season or the photograph is old.
• Keep the dated call log going, with facility, date, the name of who you spoke to, what they said, and the date you agreed to call back.
• Do not close a microchip record while the animal is only presumed. That registration is the mechanism by which a shelter you have never heard of reaches you.
• Ask each agency what species it actually impounds, not only how long it holds them.
• Ask each facility for its hold period as a calendar date rather than a number of days, and get the answer in writing.
• When a temporary shelter announces demobilisation, be there in person and get the name of whoever takes custody of unclaimed animals.
• Pick a re-check rhythm you can hold indefinitely rather than one you will abandon, and put it in a calendar with a repeat, so it survives a bad week.
• Hand the call list to somebody else on the weeks you cannot do it. It is a list of names and dates; it does not require knowing the animal.
• If a hold has expired and the animal has been placed with someone else, that is a legal question rather than a counter question. Keep every dated record and speak with an attorney licensed in your state; if you cannot pay for one, our page on what to do when someone has adopted your pet lists the free and low-cost civil legal help routes with each operator’s own stated limit.
• If any of this is hard to face, the three 24-hour lines are at the top and bottom of this page, and the two pet-loss lines with their published hours are in the section above.

What We Could Not Establish

Stating the gaps is part of the answer here, because this is a subject where a confident-sounding page is a warning sign.

No disaster-specific recovery curve was located, for any species. Described above. It is a report of what our searches returned on August 20, 2026, not a finding about the state of all knowledge.

Two hosts refused our automated requests. petcolove.org and the Petco Love Lost help center’s web front end both returned HTTP 403. The help-center article text was retrieved at HTTP 200 through that same host’s public content API, which is the route quoted here. Separately, the publisher-of-record website for the Huang study returned HTTP 403; the full text was read at PubMed Central instead, which serves the same open-access article. In both cases one route failed and another worked. Neither source is dark.

The Lord 2007 papers were read as abstracts only. Both are behind the journal’s paywall with no PubMed Central deposit. Every figure quoted from them here is in their published abstracts, and nothing beyond an abstract appears on this page.

The 800-cat Lahaina figure has no operator-page confirmation, and no period attached. Maui Humane Society’s own fire-response page, opened August 20, 2026 at HTTP 200, publishes no numbers, though the organization does publish Asilomar Accords shelter statistics by fiscal year on a separate page of its site, which we did not mine for this figure. The figure is attributed here to a named MHS officer speaking to a news outlet, which is what it is. The interview does not say over what span those 800 cats were retrieved, so this page does not put one on it.

Butte County Animal Control’s two-week statement is documented in one place. It appears in Rachel Zarnt’s first-person account and we did not locate a contemporaneous Butte County record of it. It is published here strictly as what Zarnt reports the agency said, not as a Butte County position.

We did not re-open the Marshall Fire paper for this page. Its figures live on our report-filing page, which records reading it, and we linked there rather than quoting a live page of our own as though it were a source.

Nobody has published the join this page makes. The reading that ties an ordinary-loss curve, a set of disaster rates and a stray-hold rule together into the instruction to watch the paperwork clock rather than the animal clock is this site’s, assembled from the documents above. It is labelled as ours wherever it appears, and you should weigh it as one site’s reasoning rather than as a finding.

The Crisis Lines Again

These three lines are open 24 hours a day, every day. Every number and phrase below was read on that organization’s own website on August 20, 2026. SAMHSA’s Disaster Distress Helpline page, marked Last Updated: 09/26/2025, says “Call or text 1-800-985-5990.” Spanish: “Llama o envía un mensaje de texto 1-800-985-5990 presiona “2.”” For Deaf and Hard of Hearing ASL callers: “Please text or call the Disaster Distress Helpline at 1-800-985-5990 using your preferred Relay provider.” SAMHSA states “There is no need to give any identifying information when you contact DDH.” It adds that any basic questions a counselor asks at the end of a call are optional. The 988 Suicide and Crisis Lifeline: call 988, text 988, or chat at 988lifeline.org, with the site stating “The 988 Lifeline is available 24/7/365. Your conversations are free and confidential.” The Veterans Crisis Line: “Dial 988 then Press 1”, “Text 838255”, or chat online, with the site stating “You don’t have to be enrolled in VA benefits or health care to connect.” Talking to a licensed mental health professional about any of this is a reasonable step, and it does not require calling a hotline first.

The two pet-loss lines, with their published hours and Cornell’s own not-a-mental-health-hotline limit, are in the section above.

Where to Go Next

If the search is new. Start with our post-disaster reunification workflow, which covers the first forty-eight hours in the order that gets the best odds, and our page on filing a disaster lost-pet report, which covers what goes in the report and the two clocks that run on it afterwards.

If the local search keeps coming back empty. Read what happens when a lost pet is transported out of the county or the state. An empty radius is a reason to widen the reporting, not a result.

If you are working a burn scar or a flood zone in person. Our guide to a feeding station and a trail camera for a displaced cat is the method that keeps working when the landmarks do not, and the flyer and shelter call log is the record that keeps a long search from repeating itself.

If you have found the animal and cannot get it back. Proving a pet is yours and working the reclaim process goes jurisdiction by jurisdiction, including what happens when a hold has already expired. If the animal has already been adopted by somebody else, what to do when someone has adopted your pet covers that, and it carries the free and low-cost civil legal help routes with each operator’s own stated limit.

If another animal in the household is still with you while this one is missing. The surviving pet when a housemate is missing is written for that household, and this page does not cover it.

If you are the person who found somebody else’s animal. What to do when you find a pet after a disaster covers scanning, reporting and the duties that come with holding an animal that is not yours.

If the search has ended. What to do if your pet dies in a disaster covers the decisions in the first hours, the surviving animals in the household, and the grief routes. And guilt after leaving a pet behind is the page for the specific version of this that starts with the evacuation rather than with the search.

Then do the one thing that changes the odds for next time, and it takes ten minutes on a quiet afternoon: open the microchip registry, confirm the record names you, confirm the phone number is one you will answer from a hotel two states away, and add an out-of-area second contact. Our microchip registration audit walks it. Ellie’s collar tag was accurate and it was a landline in a house that had burned down.