By

Mark Rounds

John Shovic Ph.D

The SADL Laboratory

Department of Electrical Engineering and Computer Science

Washington State University

Pullman WA, 99164-2752

Phone: 509-335-6486

e-mail: mrounds@eecs.wsu.edu

The basic philosophy of the National Park Service is to establish "pleasuring grounds" for the "enjoyment of the people". They are also charged with preserving ecosystems and wildlife, and maintaining the remnants of our wilderness past. No one animal represents the conflict of these interests better than the Grizzly Bear (Ursus Horribilis). This animal is the most dangerous predator in the Yellowstone ecosystem, yet the very nature of the national park system puts these magnificent beasts in regular contact with human beings. Predicting the potential of interaction between the Grizzly and man would be of great value to the park service and the public at large.

During the latter portion of the summer, a large part of the Grizzly's diet consists largely of army cut worm moths and the caterpillars that precede them. If you can predict the habitat of these moths and caterpillars, you can predict the areas to which Grizzly Bears will migrate in the later portion of July and August.

This is also the time the white bark pinecones are available. The seeds of these pine trees also make up a large portion of a Grizzly Bear's diet. By looking for areas where the white bark pine and moth habitat are in close proximity, you can more closely predict Grizzly habitat in the summer and fall months, prime tourist months for the park.

Using a rule based approach for developing high probability moth habitat areas and white bark pine stands this paper will develop likely areas of bear habitat. These GIS databases will be overlaid on various human activity maps utilizing evaluation tools in ArcInfo to predict potential areas of Human/Grizzly interaction. These predictions will be validated using Grizzly sightings during the time period in question.

The Grizzly Bear¹ is the largest carnivore in the lower 48 states and one of the largest in the world. Its color varies from black to blonde, frequently with white-tipped fur, giving a grizzled or "silver-tipped" appearance. Males, on average, tip the scales at between 215 to 715 pounds and stand 3.5 feet at the shoulder, but there have been some specimens that have weighed in over 1,000 pounds. Females tend to be smaller, averaging between 200 and 400 pounds. Life expectancy is between 15 and 20 years in the wild.

The population of Grizzly Bears in Yellowstone National Park is not large, between 250 and 300 bears. The size of an individual bear's range varies from 70 square miles to over 1,000. This is based primarily on the food content of the range. Though they are often called meat eaters, carnivore is actually an inaccurate description of its dietary habits as it will eat almost anything and as such fits in the category of omnivore. This is a very important point to this study. Grizzlies rarely hunt. They forage and consume much more plant matter than meat. As a result they must consume in excess of 40 kilograms of food per day.

Although not true hibernators, both grizzlies and black bears den up during the long winter months in Yellowstone Park. 61 percent of Grizzly Bears in Yellowstone den on north slopes at elevations from as low as 6,500 feet to more than 10,000, but most are dug between 8,000 and 9,000 feet. 50 percent of all grizzly bear dens are dug under the roots of a tree that tends to give the roof support. The other 50% are dug on open hillsides.

Preparing a den typically occurs in late summer while the bear is in hyperpagia (mass feed) to build up fat. Bears are very secretive of their den sites and will abandon the site if disturbed.

The triggers to enter the den are a combination of the first heavy snow, a reduction in the supply of high-quality foods, decreased mobility due to snow, and increased energy costs of keeping warm. Pregnant sows will den first, older males last.

The trigger to exit the den is snow melt. Large adult males and single or poorly conditioned bears will exit first and then sows with cubs of the year (newborn) last. The bear is typically lethargic and will lay around on the "porch area" of the den for some time, and may re-den if winter conditions return. If warm weather continues, the bear will seek out water, or eat snow, roots and herbs to clear kidneys and digestive system, and then any winter-killed carcasses, or, more recently, wolf kills.

The physiological aspects of a hibernating bear are rather unique. They do not eat, drink, defecate, or urinate during hibernation. The digestive system and kidneys shut down almost completely. The bears exist on foods and fluids stored in their bodies. Poisonous wastes and byproducts are broken down and reabsorbed; urine is reabsorbed through the bladder wall and processed into amino acids and protein.

Females give birth while hibernating. Gestation begins at the start of denning and is six to eight weeks long. Bears mate in May or June, and through a delayed implantation, gestation is delayed until the sow enters its den. Recent research points to the sow's ability to give birth to cubs from different boars. Females mate with many males. The average litter consists of an average of two cubs, sometimes three, and rarely four. Grizzly sows keep and raise their cubs for 2 1/2 and sometimes 3 1/2 years.

It is the time during the late summer when the bears are in the hyperpagia phase that is of most interest to this study. During this time the bears will range some considerable way looking for food that includes everything from carrion (bears will only rarely kill prey) to insects to many forms of plant material.

The Grizzly Bear is a farmer's friend when it comes to getting rid of army cutworm moths. Grizzlies eat the moths in large quantities, especially in years when the bears can't find enough huckleberries.

During the month of August, each of these moths has about 0.5 calories of food value in fat, which is easily digestible. Since grizzly bears eat 20,000 to 40,000 moths in 24 hours, they're getting a significant number of their calories from moths. In a single month, the grizzly bears can devour at least 300,000 calories from moths, which amounts to about one-fourth to one-third of the total number of calories they need to live for a year.

A bear can spend almost 70 percent of its day eating moths. They find them on rocky alpine slopes that heat up quickly in the summer sun. Temperatures can reach 120 degrees. In ideal conditions, up to several hundred moths in a square meter can be found.

The University of Manitoba², in Canada has an ongoing research effort into these insects and is the source of much of the cutworm data used in this paper. Cutworm larvae have four sets of abdominal prolegs and curl up when disturbed. Army cutworms are pale greenish-gray to brown in color. They have pale stripes down the back and a mottled pattern. They also have a lighter band along the sides.

Cutworm moths may lay several hundred eggs on their host plants. After the eggs hatch, the larvae feed on the host plants. They molt several times, eventually reaching about 5 cm (2 in.) in length. The larvae tunnel into the soil to form earthen cells where they pupate. The new moths emerge, exiting through the soil using the old larval tunnels. Some species overwinter as eggs (e.g., the red-backed cutworm); others, as larvae or pupae.

The life cycle of the Army Cutworm moth is as follows.

May - Overwintering eggs begin to hatch

June - Larval feeding begins

July - Larval feeding continues, pupation begins

August - New moths emerge, lay eggs

September - Eggs overwinter

The army cutworm feeds on the foliage of wheat, oats, barley, mustard, flax, alfalfa, sweetclover, field peas, cabbage, sugar beets, corn, oats, potatoes, various weeds (notably stinkweed) and grasses. Almost any crop, present during the early spring, could be a potential host.

The white bark pine, also known as Scrub Pine, is of particular interest. During the late summer and early fall, the seeds of this tree are a prime food source for Grizzly bear³. The name 'albicaulis' means the "pine with white stems" because of the white bark evident in young trees. The plant can be in the shape of a small tree with a rapidly spreading trunk and broad crown or in the shape of a shrub with wide-spreading crown and twisted, gnarled branches. It is distinguishable from Limber Pine by the cones and the hollows around the roots are used as shelters by both animals and hikers.

The needles come in bunches of five and are three to nine centimeters long,. They are slightly curved, stiff, and colored a bluish-green. The bark in young trees is thin, smooth and chalky white. This gives the tree its distinctive name. Mature trees tend to have a thick, darker bark with ridges which is much more like the standard bark we are used to.

The cones themselves are egg-shaped to round and three to eight centimeters in length. They tend to grow at right angles to the branches and remain closed on the tree and seeds are only released when the cone decays on the ground seeds are wingless and about one centimeter long. They are rich in oil and an excellent food source.

As you can see in the table on the above, these moth pupate in the month of August. At that time, they have the most calorie content. This is also the time when the moth is most vulnerable to the bear⁴. The moths are by nature, nocturnal and during the day seek rocky soils to hide and sleep in until the evening comes when they go out to find mates, lay eggs and find flowers for nectar.

During the day they tend to hide in warm, rocky alpine soils and when they are disturbed, they tend to be lethargic in attempting to escape. There are several reasons for this behavior. First, they do sleep and have just awakened. Secondly, they are usually under a rock with limited light and the bear usually turns over the rock exposing the moths to light suddenly, which may stun them. These factors allow the Grizzly Bear time to consume the majority of those found.

The moths tend to cluster and on a given hectare of land they maybe several hundred thousand moths which is enough to feed a bear for several days. Thus even the smallest section of moth habitat has the potentiality of having bears feed.

The White Pine Bark tree is also the most vulnerable to the bear at this time. This is the period when the seed contain the most food value and are sought after by not only black and grizzly bears but also several species of bird⁵.

Because these moths tend to collect in a specific type of soil that is relatively uncommon, they can be used as a predictive measure for bear migration. Dr. Henry Shovic, who works with the National Park Service, is an expert on the soils of Yellowstone National Park and has stated that the soil type that is most attractive to Army Cut Worm moths is called Talus⁶. This is the rocky rubble that you see at the bottom of most bed rock outcroppings. With his guidance, I developed the following criteria for soils selection:

1. The soil map object must have 20% or more of Talus Slope and be largely unforested
2. Or must have 30% or more Talus slope with forest but not predominately (Talus Slopes are usually unforested)

Starting with the soils map and using these criteria, I was able, utilizing Dr. Shovic's book, Soils of Yellowstone National Park⁷ to come up with a list of the following soil types and then selected from this list further to insure they had enough slope to form Talus:

Soil Identifier Description

2154 Greyback Family-Shadow Family-Sedimentary Bedrock Complex

2422F Lolo Family-Bedrock Outcrop -Pesowyo Family Complex

2546 Hobacker Family-Bedrock Outcrop-Arrowpeak Family Complex

513 Bedrock Outcrop-Como Family Complex

5217F Bedrock Outcrop- Pesowyo Family-Lamedeer Family Complex

522 Bedrock Outcrop-Hobacker Family Complex

5294 Bedrock Outcrop-Hobacker Family-Badwater Family Complex

544 Bedrock Outcrop-Arrowpeak Family Complex

835 Josie Family-Whitecross Family-Bedrock Outcrop Complex

8357 Trude Family-Whitecross Family-Bedrock Outcrop Complex

8853 Trude Family-Bottle Family-Bedrock Outcrop Complex

When these soils are located on the Map of Yellowstone, you can see that they spread throughout the park, but tend to cluster in certain areas. These graphics were compiled using the Grid Subsystem of the ArcInfo Package and then displayed using ArcView. The cells are 93.135 meters on a side. This odd number was required because the elevation model could not be rescaled on the system available to me and so I had to rescale everything else to fit it. The elevation model was critical to the study for developing slope and aspect.

Soil grouping have been selected for soil type and the fact that they have enough slope to drain properly. The moths do not tolerate moist areas so any area that has the potential of attracting the moths are, by

definition, open and dry. This initial area encompasses over 1,383,948 hectares of the 8,893,870 hectares contained in the park. The green areas in the graphic below depict these areas.

The next graphic contains these same areas but as the moth likes warm areas, this takes into account the aspect of the terrain. Any otherwise suitable slope (that include proper soils composition and slope) that faced in a northerly direction was disqualified from further consideration. This area was 1,089,513 hectares, reducing the prime area by 294,434 hectares. The areas in green are the areas with suitable soil types and slope. The areas in brown are those areas that also have proper aspect.

To validate this study, I have received from the Park Service some generic Grizzly Bear sighting data. The perfect type of data would have been for the month of August over successive years. Unfortunately, what I was able to get was a compiled list of about 100 sightings. They were in no particular order and the Park Service did not include any dates with these sightings, just UTM locations.

I asked why this was so and was told that this is very sensitive data and real detail on Grizzly Bear habitat and migration is not available to the general public. This data can really give only a confirmation of potential habitat and cannot be used for any other purpose. Indeed the source of this data made very sure by format and the withholding of the time factor that it couldn't really be used in any other research. If you look closely at the map below, you will note that the orange flags that indicate a Grizzly Bear sighting tend to cluster along the roads.

This proves that most of the people who see bears, usually do it from their car or on the established trails. This does not indicate that Grizzlies hang out by the picnic tables looking for unwary passersby. They tend to leave humans alone when they have the opportunity to do so.

The important thing to note from this map is the fact that the large majority of sightings were in areas near moth habitat. In areas where the moths do not cluster the number of sightings is quite small. This area is some of the most rugged and unforgiving terrain in the park.

This data does show some clusters that are abnormally high. Specifically in the north west quadrant of the park. High traffic alone is not enough account for the number of sightings. This apparent conflict caused some further discussion. The following map was constructed.

From this map, it becomes apparent that the combination of a high number of campsites, hiking trails, significant white bark pine stands and the moth habitat in this area, that the potential for human/grizzly interaction goes up significantly.

This also makes clear why, even though in the south west corner of the park where there is significant moth habitat and significant camping and hiking activity, the number of bear sights in significantly lower. There is a very limited white bark pine stand in that area. IT appears that unless all three of these factors, moth habitat, white bark pine stand and human concentrations exist, bear sightings will be dramatically less.

There are some factors that need to be considered when taking any further steps in this sort of research. First, the White Bark Pine crop was especially poor for three years followed by one good year. Despite that, the bears in Glacier were four weeks late in eating moths. This could be because the summer was unusually hot and long and they didn't get the signal to start storing up fat for the winter until almost September, but we don't really know. This method would also not be as good a predictor of bear movements in years that have white bark pine crops as the food value in the pine nuts is higher and the bears seem to prefer the taste.

This study predicts that the roads through the northwestern third of Yellowstone Park lie in close proximity to Moth Habitat areas. This portion of the park should also see a high proportion of grizzly sightings which is indeed the case from the limited sighting data available. Based on this information, heightened awareness and more vigilant patrolling during the moth season in the areas described in this paper is warranted for this portion of the park.

As always, GIS analysis is not definitive. What I have done is to predict where Grizzlies might be during a given time of the year if the weather conditions are right and if the white pine bark pine nut is average or worse. It is left to some dedicated field observer to hike into this rocky wilderness during the month of August, set up a blind to watch and study and report what he or she finds. The author of this paper leaves that to a younger and more idealistic researcher. This evidence is exciting however, and does merit further study into the foraging habits of the grizzly.

I did learn some valuable GIS lessons in the scope of this project. I hope that they will be of value to future researchers.

1. Scale is important. On the first time through using this data, I did not get many small areas of prime habitat through the selection of an improper scale and only by reprocessing did they become noticeable. I also learned that even with a 400 MHz dual Pentium processor there are limits. The DEM would only scale down to 93.135 meter cells. Attempts to change the scale took four hours and ended in a locked system. I have since learned that this is because the system loads up with temporary files and utilizes all available temporary storage.

2. The Forest Service will give you exactly what you ask for, not what you need, so carefully word your requests to get the appropriate information to avoid multiple requests and angry federal employees.

3. Soil is complex. I had no idea that there were so many different soils types and their effects on the habitat. Because soil is complex, the number of different soil objects is truly immense and only careful management of my computer resource made this possible.

4. When dealing with large models, Raster (in this case the Grid portion of ArcInfo by Esri) is indispensable for developing analysis quickly. On the other hand, ArcView (Esri other product) is very handy for developing graphics for this paper and my presentation. You really do need both.

5. Manual digitization of data is not trivial and I now know why they have prisoners doing it.

6. After much searching around, I was able to locate all the GIS data with all the meta-data in good formats from a small company in Idaho called TriGeo⁸ save the bear sighting data which came from the National Park Service. Their (TriGeo's) CD is a must for anyone doing research in or around Yellowstone National Park.

7. A more complex Bear Feeding Model will lead to a more accurate predictor.

The next step in this research, now that the late summer feeding model seems to be robust is to build similar models for other times of the year. This kind of data could provide a more reliable predictor of bear and human interaction.

"The Bears of Yellowstone," by Paul Schullery, 3rd revision. Published by High Plains

"The Great Bear Almanac," by Gary Brown. Lyons and Burford

"Soils of Yellowstone National Park" by Ann Rodman, Henry Shovic and David Thoma

"Soil Survey of Gallatin National Forest, Montana" Published by the US Department of Agriculture

Website http://www.gov.mb.ca/agriculture/insects/fad06s00.html for details on the Army Cutworm Moth

Website http://www.montana.edu/wwwpb/moths.html is the source of detail on the Moth Bear connection