The USGS Investigates Elk

Guest Post by Willis Eschenbach

Elk are one of the largest of the “Cervidae”, the deer family, and are one of North America’s largest mammals. Cows weight about 225 kg. (500 pounds) while bulls weigh about 320 kg (700 pounds). They are magnificent animals in the wild, and in addition they have another very important feature. They are … well … umm … no socially acceptable way to say this in a world containing vegevores but to say that elk are delicious, as many wolves or mountain lions can testify. As can I. This has not worked to their historical advantage.

Originally there were elk over much of the US. However, because of the “delicious” factor, many of the species of elk were turned into elkburger by short-sighted humans of the melanin-deficient variety. The situation in the southwestern state of Arizona is fairly typical. There, to use the lovely language of biologists, the local elk species, Merriam’s Elk, was “extirpated” just prior to 1900.

Arizona is usually thought of as a hot desert state. But the northern part of the Arizona is mountainous. I rode a freight train through northern Arizona in winter one time, and I hope to never be that cold again in my life. But the elk didn’t mind the cold, at least until we ate them all.

However, everything is not lost, humans can also repair mistakes or at least ameliorate their effects. Groups of a different species of elk were imported from Yellowstone Park from about 1910 to 1930. At first the herds were small, but nature is nothing if not fecund, and we’ve killed a lot of the elk’s natural predators (bear, wolf, mountain lion), so the main problem now is to keep the herd size down. And it is not an insignificant problem. The size of the elk herds is putting pressure on a variety of resources, both natural and human, all over the state. The Arizona Department of Game and Fish (AZDFG) has an elk management plan that lays out all that it is doing to keep the elk numbers down, and they have a big job on their hands. Their methods include regular hunting seasons; special hunting seasons; hunting seasons designed to drive elk out of a specific area; “antlerless” hunts for female elk; special permits for farmers to shoot elk that are eating their crops; hunts designed to preserve winter forage for the winter, and the like.

This is a recurring problem with large herbivores all over the planet, particularly where we’ve killed the large carnivores. There’s not a lot of food in your average blade of grass. A big animal like an elk has to eat lots and lots of vegetation just to stay alive, and even more to gain weight. And this, of course, means that wherever elk go, they will change the local ecology big-time. A herd of elk is natures mowing machine, only they’ll also mow down small trees.

Why is all of this of interest to climate science?

Figure 1. Snowfall in Flagstaff, Arizona. Pale red squares show the 17-year centered Gaussian average of the data. The airport gets more snow than the town, and as a result I have adjusted the town data so that their averages agree during the period of overlap. As you can see the match is good. The black bar shows the 95%CI for the error of the Gaussian average at the boundary of the dataset. DATA SOURCES: “History of Flagstaff“,   Flagstaff Airport  Town Adjusted 

Well, some University of Montana folks, along with the US Geological Survey, have just published a paper called “Climate impacts on bird and plant communities from altered animal–plant interactions”, by Thomas E. Martin and John L. Maron, paywalled of course. (“Impacts2011). The authors make the  claim that human-caused climate change in the form of reduced snowfall around Flagstaff and in other mountain areas of Arizona is allowing the elk population to graze higher in the mountains in winter, and as a result the local ecology is changing.

This, of course, brings up several related questions:

1. Does an elk eat in the woods?

2. Has the snowfall in the Arizona mountains gone down lately?

3. Are there other factors that might push elk up into the mountains?

4. What other parts of Arizona are the elk moving into?

Question 1. Yes, elk eat in the woods, a herd of elk is a tree-trimming and mowing machine par excellence. But as is far too common in these kinds of studies, the authors can’t resist gilding the lily. Here are their photos showing what the elk can do …

Figure 2. ORIGINAL CAPTION e,f, Photos showing the decline in understory plant density in the same area of a study drainage from May 1985 (e) to May 2011 (f).

Now, I’ll buy that elk can do that kind of damage, because elk do eat in the woods, and they eat most everything. But I won’t buy that those photos are taken from the same location. They are careful to say that they are in the “same area”, but they are presented as a “before and after” combination, when they are nothing of the sort. For all I know, photo “f” may have looked like that for the last quarter century. A small point, I know, but that kind of thing rubs me the wrong way.

In any case, it is obvious to anyone who has been around them that elk eat in the woods, that they eat a lot, and that trees and their inhabitants suffer as a result. Bad elk, buncha cervine eco-criminals. Or as the authors say:

We excluded elk from one of two paired snowmelt drainages (10 ha per drainage), and replicated this paired experiment across three distant canyons. Over six years, we reversed multi-decade declines in plant and bird populations by experimentally inhibiting heavy winter herbivory associated with declining snowfall. Moreover, predation rates on songbird nests decreased in exclosures, despite higher abundances of nest predators, demonstrating the over-riding importance of habitat quality to avian recruitment.

So the authors have proven that yes, elk eat in the woods, and yes, they eat trees and understory of all kinds, and yes, when the elk do that, songbirds suffer. I would not have thought that it would take a scientific study to establish that, but I suppose it is good that they did. I would note in passing that one man’s ceiling is another mans floor, and if songbirds suffer, surely some other creature gains from having cleared out understory, and that includes fire protection for all the forest animals … but I digress. According to AGw supporters, climate change can only bring negative outcomes and no benefits, it is well known.

Regarding the second question, I see no evidence of any unusual decrease in snowfall. They do a typical AGW thing in Impacts2011, in their Figure (1a, not shown) they show only the snowfall from 1985 onwards. As you can see in Figure 1, 1985 was somewhere near the peak of the recorded historical swings. Overall, there is no such sign of decline in snowfall. There is a slight upward trend in the Flagstaff snowfall since 1948, but the trend is not statistically significant, nor is the level of modern snowfall historically unusual.

In addition, they provide absolutely no citation for their snowfall numbers. The closest that they come is when they say:

Snowfall at our study area has declined over the past 25 years (Fig. 1a), typical of what has occurred across western North America and other mountain regions of the world 1,7,8.

Reference 1 is the IPCC Bible, except of course they have neglected to give us the chapter and verse of the sermon. Heck, they don’t even say which volume contains the information, whether it is Working Group I, II, or III.

They are saying that supporting evidence for their claim is somewhere in the thousands and thousand of pages of the AR4 report somewhere, and by gosh, it is the reader’s responsibility to ferret it out. I despise this type of citation, and it is all too common in climate science. If I had tried citing a thousand page document with no page numbers in high school, much less college, my teachers would have had me for breakfast. Yet the authors are college professors, and the reviewers stay schtumm and don’t inquire too closely into the antecedents of the “facts” inhabiting the report.

The only good news? If someone does that, if someone puts the entire IPCC report as the very first citation for their paper without a page citation, you can rest assured that they are activists, not scientists.

Reference 7 for declining snowfall is “Effects of temperature and precipitation variability on snowpack trends in the western United States“.  You’d think with a title like that they might actually look at the snowpack trends. Instead, as the Abstract to that paper says:

In this study, the linear trends in 1 April SWE [snow water equivalent -w.] over the western United States are examined, as simulated by the Variable Infiltration Capacity hydrologic model implemented at 1/8° latitude–longitude spatial resolution, and driven by a carefully quality controlled gridded daily precipitation and temperature dataset for the period 1915–2003.

Here’s what they did. They took some real snowfall and temperature data. Then they “gridded” it, that is to say averaged it by gridcells on the map. Then they adjusted the temperatures for altitude using the moist adiabatic lapse rate. Oh, they also adjusted the precipitation based on the topography. They do not say whether they have infilled cells which contain no data or how they handle missing data

Then they’ve used that averaged, adiabatically compensated, gridded, topographically adjusted, and perhaps infilled dataset to drive a model, a model which outputs using a much smaller grid than the input data, a grid of about ten miles on a side. The paper is about the results of that model.

Sorry, not impressed. I’ve written too many computer programs, I have a good idea of the errors inherent in that process.

Reference 8 for decreasing snowpacks is “Attribution of Declining Western U.S. Snowpack to Human Effects“ It was written in 2008, yet it only uses data up to 1999. This is a huge red flag for me. The reason, in this case, is that it is the usual attempt to have climate models place the blame on humans, and the model runs end in 1999. That’s convenient for alarmism, it turns out, since snowfalls have increased in the last decade. My favorite line from that one is from the abstract, viz:

Estimates of natural internal climate variability are obtained from 1600 years of two control simulations performed with fully coupled ocean–atmosphere climate models.

I had to rub my eyes at that one, natural variability obtained from models?? I looked further, to find that’s what they mean:

… Only if changes are both outside the likely range expected due to natural climate variability and consistent with the changes ex- pected due to anthropogenic forcing can it be concluded that human activity has a role in reducing winter snowpack.

We use 1600 years of control run data from fully coupled global general circulation climate models (GCMs) to provide estimates of natural internal variability. 

And how do they determine the “human fingerprint”? More computer runs, this time with GHGs and aerosol and the whole Cirque Du Soleil. They go on to explain that step.

Multiple ensemble members of two GCMs run with estimated historical changes in well-mixed GHGs, aerosols, and ozone supply the expected response of snowpack to these anthropogenic forcings. We statistically downscale the GCM results to 1⁄8° resolution then use the downscaled fields as input to a fine-resolution hydrological model. The hydrological model calculates the SWE values as well as soil moisture, runoff, and other variables in the hydrologic water balance used in companion work

So first they are using climate model control simulation runs to give an estimate of “natural” internal climate variability, so they can rule out natural fluctuations as a cause of the change in snowfall … these guys would be hilarious if it didn’t cost us so much time and money to fight this nonsense.

Then they compare that “natural variability” to “anthropogenic” climate model runs with greenhouse gases and aerosols and the lot. To do that, they “downsize” the results of the climate models. These are typically on 5° grids or so, pretty large. They divide each gridcell into no less than a hundred little “mini-gridcells”, and “constrain” the numbers in the downsizing process by comparing them to local observational data. (As far as I know, there is no evidence that this process does better than chance regarding preciptation … but I suppose that doesn’t matter in any case, because in any case the GCMs being downsized are known to do no better than chance on precipitation.

Those thousands of mini-gridcells are then used as input to yet another model. This final model calculates the snowfall that allegedly results from the human influence … anyone care to give me some serious error estimates for that process?

Again, not impressed by the putative fingerprints. It’s models all the way down. Reference 8 is a joke, a downsizing of models known not to work for precipitation.

So their evidence of declining snowfall is both weak and out of date, as snowfalls have generally increased in the western US in the last decade. In the Flagstaff record there is nothing to suggest a human influence on the snowfall. And more to the point, current Flagstaff snowfalls are well within historical norms. During the twenty year period between the world wars, the snowfall was quite low compared to the period they studied. Makes one wonder what the elk were up to then …

Regarding the third question, of what else might push elk up into the mountains … well, duh, population pressure. This pressure comes from two sources—the elk population numbers, and the human population numbers, particularly the number of hunters. There are some issues of interest there. One is that this is not the historical elk population, which had its winter and summer ranges figured out for thousands of years. These are modern transplants whose numbers are increasing, figuring out how to survive in a plotted, parceled, complex ecoscape of natural and human forces. Are we surprised that some group of them might take to eating places they haven’t eaten before?

A second issue is that we have replaced their historical predators with human predators. One of the largest differences between the types of predation is that natural predators preferentially take the aged, the young, and the weak. Human predation preferentially takes adult males. This changes the social structure of the animals. Another difference in predation is where we hunt. Wolves hunt where they live. Humans hunt where we can get to easily, because packing out a big elk, that’s a quarter ton of meat to shlep out of the forest. So we drive the elk into and out of different areas than did the other predators. Again, should we be surprised that they are pushing up into the area of the study?

Finally, question 4. What other areas are they going into? Here, it gets interesting. The AZDFG document cited earlier says:

History and Background:

Elk did not historically occur in southeastern Arizona and are an unplanned addition to the native wildlife found there. Early elk sources such as Murie’s 1951 “Elk of North America” correctly noted that elk were not native to southeastern Arizona. However, later sources (Bryant and Maser 1982 – Elk of North America) erroneously extended the historic range of elk far in to Mexico based on unsubstantiated rumors, a report of a pictograph, and a report by Edgar Mearns’ camp cook of 2 “large deer” crossing the border into Mexico. Archaeological evidence fails to provide any evidence elk were ever in Region 5 in historic times. No evidence exists of elk remains in the fauna lists at Native American sites in southeastern Arizona.

Another large herbivore grazing on the region’s arid and fragile desert ecosystems would probably come to the detriment of other native wildlife. Elk currently occur in Units 28, 31, and 32 and can live quite well among mesquite and prickly pear. There is no doubt they would become established in many areas of southeastern Arizona and have the potential to greatly impact other native wildlife such as desert mule deer, pronghorn, and many grassland and riparian obligate species.

So in addition to elk moving up into the mountains, they are also pushing out into the desert areas.

I can hardly wait for the next study by the USGS and the University of Montana, the new study that conclusively proves that elk are moving into the desert as a result of climate change …

My best to everyone,

w.

[PS—dang science takes a while, as I was researching and writing this Anthony posted about the study here … curse you, masked man!]