Plan Your Hunt like a Democrat and Shoot Like a Republican

President Obama's resounding reelection success resulted, in no small part, to a campaign operation that used demographic information to understand and organize the american electorate in favor of the Democrats

"... That was just one of several ways that Mr. Obama’s campaign operations, some unnoticed by Mr. Romney’s aides in Boston, helped save the president’s candidacy. In Chicago, the campaign recruited a team of behavioral scientists to build an extraordinarily sophisticated database packed with names of millions of undecided voters and potential supporters. The ever-expanding list let the campaign find and register new voters who fit the demographic pattern of Obama backers and methodically track their views through thousands of telephone calls every night.

That allowed the Obama campaign not only to alter the very nature of the electorate, making it younger and less white, but also to create a portrait of shifting voter allegiances. The power of this operation stunned Mr. Romney’s aides on election night, as they saw voters they never even knew existed turn out in places like Osceola County, Fla. “It’s one thing to say you are going to do it; it’s another thing to actually get out there and do it,” said Brian Jones, a senior adviser." (www.nytimes.com 11/08/2012)

While sitting in a tree stand during this falls hunting season I had plenty of time to reflect on the sophisticated use of data used to plan the presidential campaign and wondered if similar data was available on the movement patterns of deer.   Deer density in Maine is generally less than 20 deer per square mile which equates to an average of one deer in a 32 acre parcel of land.   A successful archery hunt will require being in the right location at the right time to intercept the movement of the deer, and yet even a prime hunting location can result in zero deer sightings during the afternoon hunt.   This post explores what we know about deer movement in terms of absolute distance traveled and the expected path of the deer.   Using this data it is possible to develop a statistical picture of deer movement to compare with our hunting experience.  Productive hunting locations will be places where deer sightings exceed the random probability of deer motion due to the physical feature of the terrain.   This may seem obvious, but Romney's loss in November is clear example of the perils of not using all of the available data in planning an election or hunting campaign.  I will argue that success at putting venison in the freezer is best achieved by planning your hunt like a Democrat before you can shoot like a Repulican (OK, maybe not like Dick Cheney).

The Obama campaign used a database of human interests, the tool for a wildlife biologist is a GPS enabled tracking collar.   The collar is attached to deer and records the location of the animal every 15 minutes. Animal location data is sent from the collar to the scientist through an Iridum Satellite link at predetermined intervals.   After the survey period the collar is triggered to fall off, is recovered by the scientists, and can be reused (http://www.atstrack.com/Generic-89-More-specs.aspx).  Webb et al. (2010) used this method to track the movement of twenty two deer over seven years.  Deer locations were determined every 15 minutes providing fine scale location data and deer movement distances as a function of time of day.   Their deer motion data is shown in Figure 1 below, illustrating a clear crepuscular pattern (dawn and dusk) to deer movement.  The plot shows the average deer movement averaged for many days and many deer all plotted as a function of time of day.   The vertical lines are the standard deviations in each average.  

Figure 1.  Average hourly deer movement ploted as a function of time of day.  Notice the peaks in movement at sunrise and sunset (from Webb et al 2010). 

Male deer have the most pronounced increase in movement during the dawn and dusk, almost doubling the distance traveled during these periods relative to midday or midnight.   Interestingly, the average distance traveled by a deer is only 400 to 500 meters in an hour (about a quarter mile), well within the distance that most deer can see or hear a hunter enter a tree stand.  This means that the most successful hunter must have a quiet and protected route to the stand and plan on getting setup for the hunt well before the prime hunting times.  

Of course, a deer seldom follows a straight path through the the woods.   The extent of tortuosity of the path (curvature of the path) is an important consideration when planning your hunt since this will determine the total linear distance traveled by the deer.   An index of tortuosity is the fractal dimension of the deer movement.   A fractal dimension of ONE means that the deer only moves along a straight line.   A fractal dimension of TWO describes random motion in two dimensions typical of  the trace shown to the left http://intothefractalvoid.blogspot.com/2012/11/the-fractal-dimension.html).   We might expect that deer move in a fractal dimension close to one as they move from bedding to feeding locations and then in a fractal dimension close to two when feeding.   Webb et al. (2009) analyzed the motion of deer using the GPS trackers to determine the fractal dimension of deer movement,  and they found a very counterintuitive result.   As shown in Figure 3 below, the deer moved with the greatest fractal dimension when they were moving the fastest during the dawn and dusk.   Deer motion is the most tortuous during the prime hunting times!



Figure 3. Average distance traveled by a deer in one hour (speed) plotted as a function of time of day (Web 2010), and the fractal dimension of the deer movement (Web 2009).

Perhaps the most important implication of these data is that deer move the furthest at dawn and dusk, but not as far as we might expect because the path is very tortuous.   All of these twists and turns mean that most deer will travel much less than a 1/4 mile in an hour even at the most active times of the day.

To simulate deer motion I used a random walk algorithm to calculate a deer path as a function of time of day for a 24 hour period.  The path is calculated based on the average distance traveled and fractal dimension of motion in each hour of the day.  This is not the path of a real deer, but rather a computed path for a hypothetical deer on a hypothetical day.     The path is constrained to simulate motion from a bedding area to a feeding area and back toward the bedding area.  



Figure 4a. Simulated deer motion run one.  The numbers on the trace is the time of day.


Figure 4b. Simulated deer motion run two.  The numbers on the trace is the time of day.

The results of the simulations show expected deer movements over a 24 hour period.   This is a model of motion, not the path of a real deer.   However, like election models that directed get-out-the-vote efforts based on survey data, this model provides insights to the hunter about the general motion of deer and allows the hunter to plan an effective hunt.  A deer travels an average of 6 km (3.7 miles) in a day, but likely covers a total linear distance of less than 3000 meters (1.8 miles).  Of course, if you don't want to live in tree stand 24 hours a day then you should expect deer to travel to your stand from much shorter distances - on the order of 1000 meters or less even if you are on a well traveled deer trail.  

Again, simulations don't tell the hunter where to find deer, but combined with scouting they are very helpful in planning a hunt.   The following is a list of hunting strategies that are consistent with this new data on deer movement.

1) Unless you are driving deer with a group, the random probability of getting within shooting range of deer is very low.   Scouting the hunting area for deer sign is essential to intercepting a deer.  Enjoy some time in the summer to walk in the woods looking for deer sign.  Test your ideas of deer motion using a game camera (they now cost about the same as a dozen arrows) to document the time that deer pass a specific location.  

2) Deer move the most at dawn and dusk.   The probability of intercepting a deer is almost twice as high at these times.  

3) Get settled in your stand well before prime hunting times and plan the approach to the deer stand carefully.   Since most deer move less than 400 meters in an hour the probability of spooking a deer is high if you too are moving close to dawn or dusk.   Deer that are well away from ear and eyeshot are also several hours away from crossing your path.  

4) Deer motion changes with season.  The rut makes males do stupid things, and we need to be smarter than the deer.   Male deer movement has a lower fractal dimension (straighter path) during the rut, presumably to increase the interaction with females.  Pay attention to the succession of the food crops in your area and be aware of the influence of rifle hunters on moving deer into archery only areas.

In summary, deer motion is remarkably local until some external force makes it change.   We often see the same deer show up on a game camera week after week and then disappear for the season.   This should be no surprise.   Given the deer densities in Maine (20/sqr mile) and the average deer travel distances (4 miles total in a very tortuous path) you should expect fewer than 5 deer per day to travel past the best stand location.  Experience says that five is very optimistic number for deer sightings, perhaps because the deer detected me before I was able to detect them.

References:  

Stephen L. Webb, Kenneth L. Gee, Bronson K. Strickland, Stephen Demarais, and Randy W. DeYoung. 2010.  Measuring Fine-Scale White-Tailed Deer Movements and Environmental Influences Using GPS Collars. International Journal of Ecology,Volume 2010, Article ID 459610, 12 pages 

Stephen L. Webb,* Samuel K. Riffell, Kenneth  L. GeeAND Stephan Demaris. 2009. Using Fractal Analysis To Characterize Movement Paths Og White-Tailed Deer And Response to Spatial Scale. Journal of Mammalogy, 90(5):1210–1217, 2009



Comments