Arctic Peregrine Falcon threatened by rising rain fall caused by climate change!”
At least, that is what the media implied when it reported on the science in a paper by three scientists from the University of Alberta and Université du Québec. Between three sources I looked at – Science Daily, Edmonton Sun and CBC, a study was conducted over a three year period on 30 Peregrine Falcons in Rankin Inlet in the Territory of Nunavut.
The gist of the media articles is that over the past 30 years, there continues to be a gradual decline in the number of Peregrine Falcon chicks surviving, despite Canada banning the poison DDT in the 1970’s which was causing massive deaths. The researchers theorised that maybe rain was the cause – the chicks are completely covered in down, so if there is any heavy rain, then the chicks will get wet and possibly die of hypothermia. The author’s concluded that whether the chicks were raised in nest boxes or in natural nests, just over one-third of the chicks died in the first 21-25 days either as a result of hypothermia or flooding of the nests. The scientists also discovered that in addition, even chicks raised in the nesting boxes and protected from the rain were still prone to dieing of starvation. The author’s concluded that storms were possibly linked to the starvation rates causing lower amount of prey, and were proposing to launch a food supplementation study to test this.
What the scientific paper said:
This information comes directly from the published paper by Anctil et. al. in the journal, Oecologia in October 2013. In 2008, the author’s did some preliminary studies to verify the nesting, hatching and vulnerable periods of the chicks. They were working on the premise that at the best of times, the density of Peregrine Falcons in the area has been as high as 29 breeding pairs, or one per 12km2. In 2009, they set up 14 cameras and in 2010 they set up 15 camera’s to monitor active nests. Their goal was to see how many chicks died as a result of rain. What do they mean by that? Well, based on their previous work and studies, they have established the following:
- The chicks all hatch at about the same time – the 2nd week in July.
- The chicks are more vulnerable to hypothermia in the first 21-25 days as this is when they can’t thermoregulate, ie. maintain their body temperature.
- They concluded there were 3 ways chick was going to die in this study: (i) death by rain, (ii) death by starvation or (iii) death by other means (parent never returned, predation, siblicide, infanticide)
- A healthy chick is fed on average, three times a die:
- For a chick to die a ‘rain death,’ they had to die during a rainstorm and to be anything else, it could be outside a rainstorm.
- To die of starvation meant it was fed less than 3 times a day and didn’t necessarily die immediately the heavens opened and poured water on the chick.
- They also studied Peregrine Falcons raised in nest boxes, and generally found the survival rate to be higher. The paper does not comment if any of the chicks died as a result of heavy rain in the boxes, but instead comments that survival rates of the last hatched chicks were better in the boxes.
Anyway, based on this, 26 chicks died during the study period and 10 chicks died in 5 of the 29 nests and affected one in four chicks when they were wet, 0r 38%. The paper does not specify which year the chicks died of rain or other cause.
The Scientists concluded that the wildlife in the Arctic has not yet been studied in detail for the effects of changing climate, but the IPCC (2007) has emphasized that the “in the context of rapid climate change, weather patterns are predicted to be strongly modified.” No mention there of whether that means increased rainfall or not in the Arctic!
More importantly, they declared “the results strongly suggest that the frequency of heavy rain (ie periods of rain where >8mmof rain falls in a day) has a much greater impact on nestling survival than the total amount of precipitation recorded during the rearing period.”
Just Question It:
Let’s start questioning the media science, and to do that, having access to the original paper, we can try and work out if the media have sensatioanlised the author’s work or was it all sound science?
(1) Do the 14-15 breeding pairs of Peregrine Falcons studied in 2009-10 in the Rankin Inlet represent Peregrine Falcon populations throughout Canada? That is hard to say… It goes without saying the Peregrine Falcon numbers dropped precipitously in the 1960s and 1970s due to the use of DDT. It’s not so certain that the numbers in Rankin Inlet dropped.
However, once the use of DDT was eradicated in North America, a captive breeding program was put into play and the numbers of Peregrine Falcons have climbed steadily ever since. According to Canada’s Government Science website, the latest numbers seem to be from the year 2000 where they had risen from nothing in 1970s Southern Canada and a few dozen only in the Yukon and MacKenzie Valley (North West Territories) to over 100 pairs in the south and 400 pairs in the north.
However, there is very little data on the Peregrine Falcon numbers for Rankin Inlet. In fact, the only research available is, unsurprisingly, by one of the author’s of the paper that prompted this article – Alastair Franke. The first paper by Franke, written in 2010, focuses on pollutants as the source of deaths for0.65 young per territorial pair between 1982 and 1989 and 2002–2009 and the second paper (according to the title) focuses on the weather when the Peregrine’s take off to migrate for the winter. So, all we can determine in the Rankin Inlet is there has been a somewhat minor persistent decline in survival rates of Peregrine Falcon chicks.
However, the success in the Yukon and MacKenzie Valley of captive breeding would rather imply that captive breeding and a reduction in the use of DDT has certainly reduced the chance of the Peregrine Falcon disappearing anytime soon, and the distribution range for the Peregrine Falcon is quite extensive across Northern Canada. But… without knowing what the Peregrine Falcon population was like before DDT began wiping them out, and how much it has rebounded in Rankin Inlet, an area that (according to the paper by Anctil et, al.) appears to have only been studied since the 1980s, it is hard to determine if DDT even affected the Rankin Inlet population to begin with, if there has been recovery followed by decline or what we are seeing is natural fluctuation.
However, in the Yukon and NWT, there are over 400 pairs along, and there is an almost equivalent amount of area in the Arctic spread across Alaska, Nunavut, Quebec and some arctic islands. 15 pairs is not much of a representative sample – it would be far better to see if other populations throughout Arctic North America were studied for affects of early summer rainfall as well.
(2) Is the Peregrine Falcon threatened by rising rains in Canada? Firstly, it has to be noted that the only rainfall we are really concerned with is that in mid-July to the first week in August, as according to Anctil et. al., the hatchlings are most vulnerable in the first 21-25 days of hatching, and after 21-25 days, they are able to thermoregulate (ie. better manage their body temperature). Due to the extreme cycles in the north, the chicks do all tend to hatch at the same time, which is the 2nd week of July.
The second thing to note is rainfall in Rankin Inlet has only been recorded since 1981, according to Weatherspark (more about Weatherspark and their sources can be found here). Below, you can see the average rainfall over the last 25 years:
Now, this graph doesn’t tell you the rainfall for the 3 weeks of critical importance, but it does tell you that overall, the rainfall increased over the 3 year period of the study, but it wasn’t out of line with other years, and indeed, there have been a number of years since 1988 where the rainfall has been 30-50% higher than that during the study period. Which in the every twisting realm of questioning things, raises the question about what was the mortality rates of the hatchlings in 2011 (when there was even MORE rain vs 2013 when the rain dropped off considerably?!).
Over an extended period of time, Environment Canada looks at ALL precipitation in Canada (both rain and snow) and indicates that for precipitation between 1948 and 2011 is increasing in the Rankin Inlet area. How accurate this is has to be questioned in itself because precipitation itself has only been recorded at Rankin Inlet since 1981, and all the data is normalised to see if its anomalous relative to the precipitation records for 1961-1990 – which was deemed ‘average’ – why? I really don’t know.
But when looking at the arctic trends for spring and summer (ie when it rains), the changes are less pronounced than in the autumn and winter (when it snows), and there has been a general flattening of precipitation amounts from 1970 to 2011… In other words, it is raining a little more in Arctic Canada, but for the period that the chicks are hatching, its been pretty steady since the 1970s, which includes the period when any deaths were most likely to be attributed to DDT poisoning than weather and through the recovery after DDT was phased out.
On the other hand, Weatherspark does give an idea of the total liquid precipitation relative to the norm at Rankin Inlet in July/August, but gives it year by year – so here it is for the study period, 2008 – 2010 (you can see the raw information for 2008, 2009 and 2010 at weatherspark yourself):
From looking at the graphs above, it looks like there was some above average (shaded part) rainfall in the critical mid-July to early August period in 2008, hardly any in 2009 and none at all in 2010. I could go on to pull up graphs from other years, but they are meaningless without knowing if what the hatchling rate is, and that is not the subject of this paper.
Finally, the Anctil et. al. paper really focuses on rain being in excess of 8mm (0.3in) on any one day. For that, I’d like to present images from their own paper. The image on the left if the total number of days the rain is over 8mm from 1981 to 2010, and the one on the right is the mortality rate of the chicks against how many days there was rain over 8mm:
Well, the first interesting thing is the days of high rainfall seem to be by month – and do not specifically focus on the 21-25 days the hatchlings are more susceptible to weather events. Why do I say that? Because, look at 2010 in the Fig. 4 above – it shows there were more than 10 days of rainfall above 8mm. However, when you look at the daily data from Weatherspark, you can see there was significant rainfall at the beginning of July AND in the last 2 weeks of August. But – there wasn’t much rainfall during the period the chicks are unable to thermoregulate!
At this point, I almost can’t bear to continue because this is a fundamental flaw in the results that should have been pulled up in a peer reviewed paper (I don’t know if Oecologia peer reviews papers or not). The immense scatter in the graphs above, the fact they tried to put a line through the first 4 points in Figure 4 which are obviously highly anomalous relative to the other years AND were also when measurements first began was enough to make me treat the results with some caution. But to now find that it didn’t even rain heavily on the days the chicks are most susceptible??? How can one accept rain as the reason the chicks died if it didn’t rain (or rain heavily) during the 21 days the chicks couldn’t thermoregulate?!
Based on this fundamental flaw, its no wonder the survival rate of the chicks was higher when there was more than 10 days of rainfall greater than 8mm (Fig. 5, above) – because that reflects 2010 and that was a year with no rainfall above 8mm in the time frame the chicks were vulnerable to heavy rain.
From perusing precipitation rates for other years, it would have been interesting if the study had been going on in say 2011 when there was significantly heavy rain in early August, or 2013 when there was higher rain in mid-July….
(3) Is one-third of the chicks dieing from rain exposure statistically anomalous for Peregrine Falcons? Again, it is hard to answer for these 14-15 breeding pairs studied because as I pointed out above, we don’t know what the population was before DDT was discovered to be affecting Peregrine Falcons, we don’t know if the numbers are still recovering in Rankin Inlet and we don’t even know what the trend has been since 1981 because the Peregrine Falcons have not been studied continuously since 1981, or 1970 or 1960…
What we do know, is that as a species, Peregrine Falcon chicks are appalling at staying alive! On average, the Peregrine Falcon will lay 2-5 eggs and 60% die! Sixty percent of the chicks die! It has to be said, in comparison, a mortality rate of just above 33% is absolutely, positively healthy for the Peregrine Falcons of Rankin Inlet! It is not uncommon for many arctic bird species to lay 2 or more eggs, but one chick will die before the summer is out. It’s a hard environment, and even in the nicer temperate zones, not every chick survives in the nest as stronger chicks will edge out or kill the weaker, and the weaker ones will die of… starvation. Nothing new there and its documented in most nature documentaries on birds.
Furthermore, there appears to be no other study of Peregrine Falcons elsewhere in the world – arctic and temperate zones – to determine if there is more chance of young chicks dieing if the daily rainfall is over 8mm.
(4) Are the chicks dieing of starvation because of decreased food? Of course, as we know from above, there was no effort made to actually correlate the death of a chick with a day of heavy rainfall, so as the only criterion was being well fed and dieing on the day of some rain, there is no connection to deaths by heavy rainfall. And since the author’s also propose that they die of hypothermia when wet – was it colder than average at the time it rained? The study didn’t even comment on this, although it would stand to reason that if it was warmer, there is less chance the chick will die if wet. So let’s take a look at the temperatures ourselves for the 2008-2010 period, courtesy again of Weatherspark:.
From looking at the temperatures over the time period, it can be seen that the temperature in 2008 were on average slightly warmer than average (the red line) with a cold spell at the end of July which correlates with heavier rain. In 2009, it was overall quite cool, although it looks like there was a warm spell at the end of July/early August with the only dip in temperatures corresponding with heavy rainfall for 2 days at the end of July – if anything, this one year rather supports their theory! 2010 appears to have been a bumper year for warm conditions, and as we already know, it didn’t rain significantly at all – excellent conditions for chick survival if their deaths are to be attributed to cold, rainy weather. However, the results in the Anctil et. al. don’t correlate the chick deaths to days of heavier rain or colder weather – just that they died when the entire month had more rain days, even if it didn’t rain much when the chicks were most vulnerable.
Was there a higher amount of chick mortality in the ‘other’ category in 2010 when it hotter than in 2009? Can’t say – the paper clearly didn’t set out to study death by heat exhaustion.But there is no doubt if they had, they might have garnered more global warming alarmist media attention for their 2 year study!
(5) How significant is the 10 deaths attributed to rainstorm? Well, to answer that, how many eggs were laid in the entire study? I don’t know – they didn’t comment on that in the article. If we assumed all 29 breeding pairs laid 3 eggs (re: these birds typically lay 2-5 eggs) and the study was run fully in 2009 and 2010, then 29 breeding pairs laid approximately 174 eggs of which 10 possibly died as a result of rain, or 5%. Oooh! In fact, given the paper does state that only 26 chicks died of anything, that is only approximately 15% of the total number of eggs laid – it does all rather point (again) to a very high survival rate for the Peregrine Falcons of Rankin Inlet.
- The bad media articles are based on bad science…
- The study did not take into account days of rainfall over 8mm during the 21-25 day period that the chicks were unable to thermoregulate but instead counted all the days in July and August that had more than 8mm of rain. This does not allow any correlate chick death which high rainfall.
- There is no comparative study of Peregrine Falcons anywhere else to see if rainfall hastens deaths of chicks
- There is no evidence based on annual precipitation rates of Rankin Inlet in the last 25 years to indicate there is a measurable increase in heavy rainfall, and in particular during the study period.
- A study of food supplementation has so many variables in it that it will most likely only reveal a higher survival rate – whether it rains or not.
At best, I’d say the author’s identified another way for Peregrine Falcons chicks to die in the arctic, which is in no way statistically related to climate change at all and sits well below death of chicks due to other causes.
So was the media wrong? Unfortunately, here I think their articles were based on bad science and they only reported what the scientific paper had said. But bad science still makes for bad media science!